Summer employee positions at the Department of Computer Science 2024

The Department of Computer Science is looking for summer employees

We at the Department of Computer Science want to offer motivated students a chance to work on interesting research topics with us. We are looking for BSc or MSc degree students at Aalto or other universities to work with us during the summer 2024. If you have enjoyed your studies and want to learn more about computer science, this might be your place. We do not expect you to have previous research experience; this could be the start of your bright researcher career! You will be supported by other summer employees, doctoral students and postdocs at the department.

Ready to apply?

See the complete list of the available topics below (will be fully updated by January 2^nd)) and choose the topic(s) (max. 5) that interest you the most. Choose them from the multiple-choice questionnaire on the application form and list them in the order of preference.

Please submit your application through our recruitment system. The application form will open on January 2^nd and close on January 29^th, 2024, at 23:59 Finnish time (UTC +2).

Link to the application form: https://aalto.wd3.myworkdayjobs.com/aalto/job/Otaniemi-Espoo-Finland/Summer-Employee-positions-2024-at-the-Department-of-Computer-Science_R38184

Are you an international student or coming from abroad?

Please check the Aalto Science Institute AScI internship programme for international summer employees.
https://www.aalto.fi/en/aalto-science-institute-asci/how-to-apply-to-the-asci-international-summer-research-programme

AScI arranges activities for international summer employees who have applied through their call and helps in finding an apartment in Espoo.

More information

If you have questions regarding applying, please contact Susanna Holma from HR Team. [email protected].

Summer employee topics 2024

Topics are listed here and will be fully updated by January 2^nd at the latest.

1. eHealth services
Professor/Academic contact person for further information on topic: Sari Kujala
Contact info: [email protected]
Number of open positions: 1

We are looking for a Master thesis worker on topics related to usability and accessibility of eHealth services from patients’ point of view. Fluent Finnish skills is needed.

2. Research in the Aalto-ML4H group
Professor/Contact: Pekka Marttinen
Contact info: [email protected]
Number of open positions: 1-2

Examples of possible topics include 1) probabilistic machine learning for healthcare, 2) foundation models and reinforement learning, and 3) causal representation learning for time-series. Other machine learning topics may also be considered depending on the skills and interests of the applicant. Examples of recent articles by the group can be found in https://users.ics.aalto.fi/~pemartti/.

Skills / prerequisites: Strong basic skills in mathematics, statistics, and programming. Studies in machine learning are considered a strength.

3. Model misspecification in simulation-based inference
Supervisor: Samuel Kaski, Ayush Bharti
Email: [email protected]
Number of open positions: 1

PML focus area: Simulator-based inference
Simulation-based inference (SBI) methods are used to fit complex, simulator-based models with intractable likelihood function to data. Such models appear in a various field of science and engineering such as population genetics, radio propagation, and cosmology. Recently, the use of neural network-based conditional density estimators have become the state-of-the-art in SBI. However, they have been shown to be susceptible to model misspecification, yielding misleading and erroneous inference outcomes.

In this project, you will investigate the reasons why conditional density estimators such as normalizing flows perform so poorly under misspecification. Students with strong background in statistics and deep learning are especially encouraged to apply.

References:
Huang, D., Bharti, A., Souza, A., Acerbi, L., & Kaski, S. (2023). Learning Robust Statistics for Simulation-based Inference under Model Misspecification. arXiv preprint arXiv:2305.15871.
Cannon, P., Ward, D., & Schmon, S. M. (2022). Investigating the impact of model misspecification in neural simulation-based inference. arXiv preprint arXiv:2209.01845.

4. Dual active learning for Human-AI teaming
Supervisor: Samuel Kaski, Ali Khoshvishkaie
Email: [email protected]
Number of open positions: 1

PML focus area: collaborative decision-making and design with AI
Dual learning [1, 2] is a machine learning paradigm involving training two primal and dual models together. The tasks hold a specific connection so that the output of each serves as the input of the other, creating a loop. Jointly training of the two models can further improve them by providing valuable signals in various ways, for instance, by labelling the output of the other. In this project, we aim to actively involve a human expert inside this loop. 
The idea is to engage the expert to provide data for a primal task, which we care about, by actively performing the dual one, which is easier. For instance, generating a molecule holding particular properties (primal task) might be too tricky for chemists; however, recognising the properties of a given molecule (dual task) would be easier. By using active dual learning in this specific setting, we leverage human knowledge to improve a difficult task by performing a less complicated one.

We are looking for students with good programming skills (PyTorch) and a background in machine learning. Knowledge of active learning is considered a plus.

Keywords: active learning, dual learning, knowledge elicitation

References:
[1] He, D., Xia, Y., Qin, T., Wang, L., Yu, N., Liu, T. Y., & Ma, W. Y. (2016). Dual learning for machine translation. Advances in neural information processing systems, 29.
[2] Qin, Tao. Dual learning. Singapore:: Springer, 2020.

5. Sample-efficient Inverse reinforcement learning
Supervisor: Samuel Kaski, Mahsa Asadi
Email: [email protected]
Number of open positions: 1
PML focus area: Multi-agent RL-based user modeling

Designing a reward function is challenging in many problems specially when a human is involved and we need to find out what reward function she/he has in mind. Inverse reinforcement learning (IRL) approaches provide solutions to this type of problems when given a policy and/or demonstrations from the expert, we try to learn the reward function of the reinforcement learning problem. One of the challenges in IRL is that they usually are very sample inefficient and in this project we would like to explore the realm of IRL and focus on the solutions that have gone towards sample-efficiency. We will focus on one state of the art approach and do implement it. Of course suggestions to further extend the project are also welcome.

Background requirement:
You should know python programming language and be familiar with reinforcement learning.

References:
[1] Ng, Andrew Y., and Stuart Russell. "Algorithms for inverse reinforcement learning." Icml. Vol. 1. 2000.
[2] Bıyık, Erdem, et al. "Active preference-based gaussian process regression for reward learning." International journal of robotics research. 2023.

6. Causal Machine Learning for Precision Medicine Applications
Supervisor: Samuel Kaski, Sophie Wharrie
Email: [email protected]
Number of open positions: 1
PML focus area: Applications

Precision medicine aims to achieve goals such as estimating an individual's risk of developing disease and assigning optimal treatments based on individual characteristics. Machine learning presents opportunities to learn from larger datasets of genetic and environmental factors of disease, but there are several challenges that affect the accuracy and generalisability of machine learning models applied to these problems. 
Medical datasets are often described as "heterogeneous", meaning that different patients may have different causes and progression of disease. It is important that machine learning models account for this heterogeneity to produce highly accurate predictions for individual patients. Recent work in PML group [1] has developed a new probabilistic machine learning technique to handle these challenges by explicitly modeling differences in the causal mechanisms of disease across patients. In this project you will work with world-class biobank and registry datasets (e.g., UK Biobank, FinRegistry) and explore extensions of this new modeling approach to improve the generalisability of machine learning algorithms for applications in healthcare and precision medicine.

Relevant skills: Python programming (PyTorch), familiarity with causality and probabilistic machine learning methods

References:
[1] Wharrie, S., & Kaski, S. (2023). Causal Similarity-Based Hierarchical Bayesian Models. arXiv preprint arXiv:2310.12595.

7. Width-Uncertainty in neural networks
Supervisor: Samuel Kaski, Trung Trinh
Email: [email protected]
Number of open positions: 1
PML focus area: Bayesian deep learning

Bayesian neural networks (BNNs) are neural networks (NNs) whose weights are represented by a distribution. Compared to a deterministic NN, BNNs theoretically can produce more accurate and better calibrated predictions. However, due to the sheer amounts of parameters in modern NNs, BNNs are difficult to train and require massive amounts of computation. To circumvent this problem, [1] proposed inferring the distribution over network depths insteads of the distribution over weights. In this project, we explore an alternative direction in which we study inferrence over the width of a network. We then combine these two approaches to perform inferrence over both depths and widths.

Relevant skills: Python programming, Deep learning frameworks (Pytorch, JAX), knowledges of deep learning and Bayesian methods.

[1] https://arxiv.org/pdf/2006.08437.pdf

8. Efficient methods for uncertainty in Deep learning
Supervisor: Samuel Kaski, Trung Trinh
Email: [email protected]
Number of open positions: 1
PML focus area: Bayesian deep learning

Bayesian neural networks (BNNs) are neural networks (NNs) whose weights are represented by a distribution. Compared to a deterministic NN, BNNs theoretically can produce more accurate and better calibrated predictions. However, due to the sheer amounts of parameters in modern NNs, BNNs are difficult to train and require massive amounts of computation. Methods have been proposed to improve the efficiency of BNNs, for instance by performing inference in the node space [1] or in the depth space [2].
In this project, we will survey the methods and applications of efficient BNNs, as well as whether or not these methods can be combine together to obtain better performance.

Relevant skills: Python programming, Deep learning frameworks (Pytorch, JAX), knowledges of deep learning and Bayesian methods.

References
[1] https://arxiv.org/pdf/2005.07186.pdf
[2] https://arxiv.org/pdf/2006.08437.pdf

9. Amortized inference for targeted Bayesian experimental design
Supervisor: Samuel Kaski, Daolang Huang
Email: [email protected]
Number of open positions: 1
PML focus area: Collaborative decision-making and design with AI

Traditional approaches in Bayesian experimental design often involve repetitive, computationally intensive simulations to infer optimal experimental settings. These methods can be significantly improved through the application of neural networks, which offer more efficient and scalable solutions.

This project focuses on developing an advanced neural network system for sequential Bayesian experimental design, specifically tailored for downstream decision-making tasks. The core objective is to implement an amortized inference framework that enhances the decision-making process for a designated task. In this project, you will create a neural network that can effectively handle sequential data for experimental design. This model should be capable of adapting to new data, facilitating continuous learning and improvement.

Prerequisites:
Deep learning & neural networks (PyTorch experience is preferred), basic understanding of Bayesian inference.

References:
Maraval, A., Zimmer, M., Grosnit, A., & Ammar, H. B. (2023). End-to-End Meta-Bayesian Optimisation with Transformer Neural Processes. arXiv preprint arXiv:2305.15930.
Foster, A., Ivanova, D. R., Malik, I., & Rainforth, T. (2021, July). Deep adaptive design: Amortizing sequential bayesian experimental design. In International Conference on Machine Learning (pp. 3384-3395). PMLR.

10. AI assistant for enzyme design
Supervisor: Samuel Kaski, Ville Hyvönen
Email: [email protected]
Number of open positions: 1
PML focus area: Collaborative decision making and design with AI

Synthetic biology is an emerging multidisciplinary field where organisms are redesigned so that they gain novel functions. As a concrete example, bacteria can be redesigned to produce silkworm silk or biodegradable bioplastic. Microbial cell factories have a potential for making industrial production more sustainable since they enable utilizing renewable inputs, such as waste streams, for production.

The goal of BioDesign project is to design novel enzymes for a desired enzymatic function (that would for instance enable using bacteria containing the novel enzyme for production of bioplastic). While there has been a recent breakthrough in the related problem of protein folding (see, e.g., AlphaFold [1]), the prediction of enzymatic function of a protein is still an open scientific problem. In addition, for any specific enzyme design problem, there is very limited data available. Thus, the design-build-test-learn (DBTL) loop cannot yet fully be automated.
Instead, our goal is to build an AI assistant that combines cutting-edge deep learning methods for molecule generation (see e.g., AbODE [2]) to human expertise. The design process will be iterative so that an AI assistant will suggest molecule designs, and the expert can accept or reject them, and/or suggest further modifications to the accepted designs. The AI assistant will learn a joint model of the molecule design task and the expert knowledge with an objective of improving the biocatalytic function of an enzyme. The AI assistant and the expert will jointly improve in the molecule generation task.

You will work on building a joint model of the expert knowledge and the molecule design task. This is a novel and challenging probabilistic modeling problem. In addition, you will learn about advanced deep learning tools for molecule generation, and work in an application that can have a significant real-world impact. The position requires knowledge of probabilistic modeling (that is, background in machine learning or statistics) and solid programming skills (for instance, in Python).

References
[1] Jumper, J., et al. "Highly accurate protein structure prediction with AlphaFold." Nature 596.7873 (2021): 583-589.
[2] Verma, Y., Heinonen, M., & Garg, V. (2023). AbODE: Ab initio antibody design using conjoined ODEs. In Proceedings of the 40th International Conference on Machine Learning (ICML), pp. 35037-35050.

11. Amortized inference for multi-agent user modeling 
Supervisor: Samuel Kaski, Alex Hämäläinen
Email: [email protected]
Number of open positions: 1
PML focus area: Multi-agent RL-based user modeling

User modeling is one of the basic elements in building AI tools for collaborative human-AI settings. The basic principle of user modeling is to enable the AI to infer relevant information about humans, such as their objectives, based on the current interaction. The inferred information may then be used by the AI to efficiently guide its behavior depending on the situation.

One of the fundamental challenges of user modeling is the limited availability of interaction data with humans in practical settings which makes learning accurate user models difficult. This project primarily considers developing user modeling techniques based on computational models of human cognition; cognitive models can often act as strong and accurate prior models over human decision making, hence significantly reducing the data requirements for obtaining useful user models. The practical application of these models in collaborative real-life settings however remains a challenge due to their computational complexity. In this project, you will contribute to the ongoing development of amortized inference methods for addressing these challenges.

Prerequisities: Bayesian statistics, deep learning, basic understanding of reinforcement learning. Good programming skills is a plus.

References:
Moon, Hee-Seung, Antti Oulasvirta, and Byungjoo Lee. "Amortized inference with user simulations." Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 2023.
Hämäläinen, Alex, Mustafa Mert Çelikok, and Samuel Kaski. "Differentiable user models." Uncertainty in Artificial Intelligence. PMLR, 2023.

12. Human-In-The-Loop Causal Discovery 
Supervisor: Samuel Kaski, Nazaal Ibrahim
Email: [email protected]
Number of open positions: 1
PML focus area: Collaborative decision-making and design with AI

Causal graphs are in general necessary to predict how treatments/policies/interventions affect certain variables in a system of interest. The task of learning the true casual graph, called causal (structure) discovery/learning, is very challenging. Even in the limit of infinite observed samples, we can only learn a set of possible graphs, called the Markov Equivalence Class (MEC), which contain the true graph. To narrow this down, the solution is to often make further assumptions (which is restrictive) or perform interventions in the system to gather more informative data (which may be infeasible/impossible in some real-world applications). One other source of data is the domain expert, who could provide information about the edge structure directly. The use of expert's knowledge alongside observed samples paves a way forward to reduce the size of the MEC learnt from observational data.

One problem however is that many causal discovery algorithms that rely on observational data alone may give varying results, partly due to the different assumptions they work on (e.g. see Figure 1 in [1]). While in [1] the authors consider navigating through the space of causal graphs, we could also integrate expert feedback about the edge structure directly by for e.g. querying the expert on whether an edge exists or not [2].

This project will involve using existing causal discovery algorithms (PC [3], GES [4], NOTEARS [5] etc) and actively incorporate expert feedback on edges in the causal graph while maintaining the acyclicity constrain on the causal graph.

Prerequisites/Relevant skills: Python programming, basics in optimization and statistics.

[1] Technical Note: Incorporating expert domain knowledge into causal structure discovery workflows, Mäkelä et al. https://bg.copernicus.org/articles/19/2095/2022/bg-19-2095-2022-discussion.html
[2] Targeted Causal Elicitation, Ibrahim et al. https://openreview.net/forum?id=oqRw-a4rf36
[3] Causation, Prediction and Search by Spirtes, Glymour, Scheines (Section 5.4.2, Page 84)
[4] Optimal Structure Identification with Greedy Search, Chickering D.M. https://jmlr.org/papers/v3/chickering02b.html
[5] DAGs with NO TEARS: Continuous Optimization for Structure Learning. Zheng et al. https://proceedings.neurips.cc/paper/2018/hash/e347c51419ffb23ca3fd5050202f9c3d-Abstract.html

13. Human-in-the-loop assisted molecular design and retrosynthesis planning
Supervisor: Samuel Kaski, Yujia Guo, Yasmine Nahal
Email: [email protected], [email protected]
Number of open positions: 1
PML focus area: Collaborative decision making and design with AI, Applications

Development of new drugs is a design-make-test-analyse cycle. Usually, the “design” (identifying novel molecules) and “make” (synthetizing them) steps are crucial for a successful cycle. The design step has been empowered by generative models of molecules that can be optimized using reinforcement learning to achieve a specific goal or objective, often represented by expert preferences as well as other physico-chemical properties. While many advances were made to improve the design of novel molecules, recent studies focused on connecting the “design” and “make” steps, for example, by considering the synthetizability or feasibility of molecules as part of the goal.

In this project, we aim to develop a closed-loop workflow with an efficient coupling of drug discovery and retrosynthesis, where human expertise plays a contributing role. By leveraging both molecular property and retrosynthetic route preferences, we design molecules that align with the goal of a given drug design project an expert is supervising, and also whose retrosynthesis planning is preferred by the expert. Specifically, we will design an active learning algorithm that balances between these preference objectives, guiding the exploration of chemical space towards a more optimal solution.

Relevant skills: reinforcement learning, deep learning, probabilistic machine learning, Python programming

References:
Sundin, Iiris, et al. "Human-in-the-loop assisted de novo molecular design." Journal of Cheminformatics 14.1 (2022): 1-16.
Genheden, Samuel, et al. "AiZynthFinder: a fast, robust and flexible open-source software for retrosynthetic planning." Journal of cheminformatics 12.1 (2020): 70.
Szymanski, Nathan J., et al. "An autonomous laboratory for the accelerated synthesis of novel materials." Nature (2023): 1-6.

14. Software Development in Public Sector Organizations
Professor/Academic contact person for further information on topic: Casper Lassenius
Contact info: [email protected], 040 3500 194
Number of open positions: 2

Software projects, often under the heading of “digitalization” in the public sector tend to be large, costly, and challenged on traditional dimensions, such as cost, schedule, and quality. Projects are either conducted in-house, outsourced to consultants, or as a combination of the two, and increasingly adopt agile software development methods.

We are looking for students interested in understanding and improving how software development is done in the public sector. Students are expected to have a basic understanding of software engineering and agile methods.

Based on the interests and language skills of the applicant, we can offer either the possibility of conducting a literature review on public sector software development, or the possibility of analyzing interview data from a large Finnish public sector organization about the adoption of agile development in in-house development. The latter requires excellent skills in Finnish.

The details of the topic and tasks involved will be agreed upon together with the selected candidates.

15. Machine learning: Foundations and New Frontiers
Professor/Academic contact person for further information on topic: Vikas Garg
Contact info: [email protected]; [email protected];
Number of open positions: Around 5

Applications are invited for various summer positions in our group. An ideal student would be eager to push the frontiers of science; have strong mathematical, theoretical, statistical, or algorithmic background; and be comfortable with programming in Deep Learning. We particularly invite students with strong Pure, Applied Math, Bioinformatics, or Physics backgrounds to apply.  We also value diversity and encourage candidates from underrepresented backgrounds to apply.

Interns in our group from previous years have produced stellar research, including, a publication accepted with Oral presentation at NeurIPS 2023.

Topics of particular interest include but are not limited to:
Generative Models;  (Temporal) Graph Neural Networks, Topological Deep Learning, Topological Data Analysis (e.g., Persistent Homology); Neural ODEs/PDEs/SDEs, Deep Equilibrium Models, Implicit Models; Differential Geometry/Information Geometry/Algebraic/Spectral Methods for Deep Learning; Learning under limited data, distributional shift, and/or uncertainty; Conformal Prediction;  (Approximate) Equivariant and Invariant models; Fair, interpretable, or explainable methods; Off-policy reinforcement learning, inverse reinforcement learning, and causal inference;  Multiagent systems and AI-assisted human-guided models;  Applications in large language models, drug discovery, material design, synthetic biology, quantum chemistry, etc.; and Quantum Machine Learning.

16. Designing Tools and Practices for Responsible AI for Public Services 
Professor/Academic contact person for further information on topic: Ana Paula Gonzalez Torres, Kaisla Kajava (supervised by Prof. Nitin Sawhney) 
Contact info: [email protected]; [email protected] 
Number of open positions: 2

With the emerging adoption of algorithmic decision-making and generative AI across societal sectors, there is a need for supportive AI policies and regulations (e.g., the European Union’s Artificial Intelligence Act), which creates many socio-technical challenges for responsible and trustworthy AI. 
We are seeking to hire a summer researcher to work in applied governance frameworks for trustworthy/responsible AI which include the application of experimental frameworks for regulation and potential implementation of Machine Learning Operations (MLOps) to support transparency, accountability, and auditability.

The research is part of the Civic Agency in AI (CAAI) project being conducted in collaboration with providers of AI-based public services in Finland. The CAAI project aims to understand citizens’ algorithmic literacy, agency, and participation in the design and development of AI services in the Finnish public sector in order to advance more democratic and citizen-centric digital infrastructures.
Applicants must show a keen interest in proposed research topics and bring a mix of technical and qualitative skills and/or interest in one or more of these aspects: trustworthy AI, programming MLOps pipelines, data science and/or Natural Language Processing (NLP), AI policy, societal impact of AI-based systems, and AI discourses.
You would join the CRAI-CIS research group in the Computer Science department at Aalto University. The transdisciplinary group explores the impact of technology in critical societal contexts, working at the intersection of computational and social sciences engaging HCI and participatory design. More here: https://crai-cis.aalto.fi ; https://crai-cis.aalto.fi/civic-agency-in-ai/

17. Probabilistic machine learning for multi-modal data
Professor/Academic contact person for further information on topic: Arno Solin (for more info Marcus Klasson, [email protected])
Contact info: [email protected]
Number of open positions: 2

We are seeking motivated and talented interns to join our current research projects focused on probabilistic machine learning with positions in tractable modelling, uncertainty quantification in deep learning, and multi-modal (computer vision + language) modelling. This project is part of our broader initiative in multi-modal modelling, aiming to advance the frontiers of understanding and methods in the field of machine learning. More specifically, the research interests are in uncertainty quantification in large-scale vision/language models, as well as combining semantic understanding with scene reconstruction (Gaussian splatting / NeRF models).

Interns will have the opportunity to work on cutting-edge research problems, including uncertainty quantification in neural networks and the development of innovative inference methods. Our team values creativity, analytical skills, and a collaborative spirit. A successful candidate is expected to have knowledge of probabilistic modelling and approximate inference, and general machine learning methods as well as experience with programming in Python (e.g., TensorFlow, JAX, PyTorch, etc.).

This internship presents a unique opportunity to contribute to significant research in a dynamic and supportive environment. We encourage students who are enthusiastic about probabilistic modelling and have a keen interest in language models and computer vision to apply. Highlight your specific skills and interests in your application to align with our team's needs.

See the supervisor’s home page for representative publications: https://arno.solin.fi

18. Machine learning for sensor fusion and vision 
Professor/Academic contact person for further information on topic: Arno Solin (for more info Marcus Klasson, [email protected])
Contact info: [email protected]
Number of open positions: 1

We are seeking motivated and talented interns to join our research team to work on an applied data capture and real-time analysis application in computer vision. The intern will have the opportunity to work with the Luxonis RAE (https://shop.luxonis.com/products/rae) robot and Luxonis cameras for data capture and scene reconstruction and analysis. The applications relate to Gaussian splatting / NeRF modelling as well as general-purpose segmentation models.

Our team values creativity, analytical skills, and a collaborative spirit. A successful candidate is expected to know about general machine learning, have some tinkering background, and experience with programming in C++ and Python (e.g., TensorFlow, JAX, PyTorch, etc.). This internship presents a unique opportunity to contribute to significant research in a dynamic and supportive environment. We encourage students who are enthusiastic about machine learning and interested in real-time modelling and computer vision to apply. Highlight your specific skills and interests in your application to align with our team's needs.

See the supervisor’s home page for representative publications: https://arno.solin.fi

19. Digitaalisten ihmistieteiden aineistojen ja sovellusten kehitystyö Sampo-järjestelmiin liittyen
Professor/Academic contact person for further information on topic:
Prof Eero Hyvönen
Contact info: [email protected]
Number of open positions: 1-2
[Finnish fluency is needed in these jobs]

Monitieteinen Semanttisen laskennan tutkimusryhmä (SeCo) https://seco.cs.aalto.fi toimii Aalto-yliopiston tietotekniikan laitoksella ja Helsingin yliopiston humanistisessa tiedekunnassa, Digitaalisten ihmistieteiden keskuksessa HELDIG. Tutkimusryhmässä kehitetään kulttuurialan kansallista semanttisen webin tietoinfrastruktuuria ja siihen perustuvia sovelluksia lukuisten yhteistyökumppaneiden kanssa, erityisesti Sammoiksi kutsuttuja datapalveluita, portaalisovelluksia ja data-analyyttisiä työkaluja. Tekeillä on esimerkiksi Kansallisgallerian ja suomalaisten taidemuseoiden aineistoja julkaiseva Taidesampo ja arkeologian aineistoihin liittyvä PASampo ja Rahasampo yhteistyössä British Museumin ja Kansallismuseon kanssa.

Tarjolla oleva työ pyritään suuntaamaan hakijoiden osaamisen ja kiinnostuksen kohteen mukaisesti ja projektoimaan harjoitustöiksi, joista saa palkan ohella myös opintopisteitä.

Lisätietoa sammoista: https://seco.cs.aalto.fi/applications/sampo/

20. Massively Parallel Algorithms for Graph Problems 
Professor/Academic contact person for further information on topic: Jara Uitto
Contact info: [email protected]
Number of open positions: 1

Parallel processing of data and distributed computing are gaining attention and becoming more and more vital as the data sets and networks we want to process are overgrowing the capacity of single processors. To understand the potential of modern parallel computing platforms, many mathematical models have emerged to study the theoretical foundations of parallel and distributed computing. In this project, we study algorithm design in these models with a particular focus on the Massively Parallel Computing (MPC) and Local Computation Algorithms (LCA) models.

The problems we study are often in (but not limited to) the domain of graphs, which serve as a very flexible representation of data. We are interested in, for example, the computational complexities of classic problems such as finding large independent sets, matchings, flows, clustering problems, etc.

The applicant is assumed to have a solid knowledge of mathematics, knowledge of the basics of graph theory, and a good command of English. No prior knowledge in distributed computing is required, although it might be helpful.

21. Deep generative modeling for electronic health records 
Professor/Academic contact person for further information on topic: Assoc. Prof. Harri Lähdesmäki
Contact info: [email protected]
Number of open positions: 1+

We are looking for summer interns to develop novel probabilistic machine learning methods for large-scale electronic health datasets from biobanks and clinical trials. This project aims to develop novel deep generative modeling methods to (i) predict adverse drug effects or other outcomes using longitudinal/time-series data from large-scale biobanks and clinical trials, and to (ii) harmonize large-scale health data sets for AI-assisted decision making to revolutionize future clinical trials. Methodologically this project can include e.g. VAEs, Bayesian NNs, domain adaptation, Gaussian processes and causal analysis. Experience/Studies on probabilistic machine learning is expected. Tasks for summer internship can be adapted to fit student's skills and interest. The work will be done in collaboration with other research groups from the Finnish Center for Artificial Intelligence, and the novel methods will be tested using unique real-world data sets from our collaborators in university hospitals and big pharma company. Work can be continued after the summer.

Our selected recent work:
[1] https://arxiv.org/abs/2311.03002
[2] https://doi.org/10.1016/j.patcog.2023.110113
[3] http://proceedings.mlr.press/v130/ramchandran21b.html
[4] https://academic.oup.com/bioinformatics/article/37/13/1860/6104850

22. Deep generative modeling for dynamical systems
Professor/Academic contact person for further information on topic: Assoc. Prof. Harri Lähdesmäki
Contact info: [email protected]
Number of open positions: 1+

There is an abundance of dynamical systems around us, but many real-world systems are too complex to be modeled explicitly. Recent machine learning breakthroughs include black-box modeling methods for differential equations, such as Gaussian process ODEs [1], neural ODEs, neural PDEs [2], etc. These methods are particularly useful in learning arbitrary continuous-time dynamics from data, either directly in the data space [1,3] or in a low-dimensional latent space if data is very high-dimensional [4,5,6]. We are looking for summer interns to join our current efforts to (i) develop efficient yet calibrated Bayesian methods to learn such black-box differential equation models directly from data (e.g. in robotics, biology, physics or video applications) using a low-dimensional latent space representation, and (ii) to further developing these methods for reinforcement learning [7] and causal analysis e.g. in health and other applications. Methodologically this project can include e.g. VAEs, neural ODEs, variational inference, reinforcement learning, and causal analysis. Experience/Studies in probabilistic machine learning and differential equations is expected. Tasks for summer internship can be adapted to fit student's skills and interest. Work can be continued after the summer.

Our selected recent work:
[1] http://proceedings.mlr.press/v80/heinonen18a.html
[2] https://openreview.net/forum?id=aUX5Plaq7Oy
[3] https://arxiv.org/abs/2106.10905
[4] https://papers.nips.cc/paper/9497-ode2vae-deep-generative-second-order-odes-with-bayesian-neural-networks
[5] https://arxiv.org/abs/2210.03466
[6] https://arxiv.org/abs/2307.04110
[7] https://proceedings.mlr.press/v139/yildiz21a.html

23. Deep learning for single-cell biology
Professor/Academic contact person for further information on topic: Assoc. Prof. Harri Lähdesmäki
Contact info: [email protected]
Number of open positions: 1+

Single-cell sequencing technologies provide genomics data at unprecedented resolution and deep learning methods are commonly used to analyze these datasets. We are looking for a summer intern to develop novel probabilistic machine learning and deep learning methods for various tasks in single-cell biology, including e.g. (i) predicting treatment responses and (ii) modeling spatial single-cell. Methodologically this project can involve various neural network architechtures (attention, GCN), generative models and probabilistic machine learning. Experience/Studies in probabilistic machine learning and deep learning as well as interest in bioinformatics are expected. Tasks for summer internship can be adapted to fit student's skills and interest. Work can be continued after the summer.

Our selected recent work:
[1] https://doi.org/10.1093/bioinformatics/btad743
[2] https://academic.oup.com/bioinformatics/article/39/1/btac788/6881078?login=false
[3] https://www.nature.com/articles/s41467-022-33720-z
[4] https://www.aalto.fi/en/news/an-ai-model-reveals-how-the-bodys-defence-system-recognises-skin-cancer

24. Research assistant for visual algorithm simulation
Supervisor: Senior University Lecturer Ari Korhonen / Doctoral Researcher Artturi Tilanterä
Email: [email protected]
Number of open positions: 1

We are conducting research to improve teaching of Data Structures and Algorithms. The study is related to visual simulation exercises developed for the course. The aim is to record students’ answers in detail to improve the exercises. See the following five minute introductory video: https://users.aalto.fi/~atilante/videos/Tilantera-CS-research-day-2021.mp4. Ultimately, your work will help students and instructors in programming courses. Your task would be developing further the software which records students’ solutions to these visual algorithm simulation exercises. Here you can see a couple of examples of the recorder software: https://jsav-player-test-app.web.app/.

Requirements: You should have completed the following Aalto University courses or a similar course in some other university:

One of these:
- CS-A1110 Data Structures and Algorithms
- CS-A1141 Tietorakenteet ja algoritmit Y
- CS-A1143 Data Structures and Algorithms Y
- Other similar course

In addition, one of these:
- CS-C3170 Web Software Development
- Experience with JavaScript programming

It is a plus if you already know Git, Node.js, and JSON Schema. In addition, any general programming skills are considered to be a merit. Please, add a link to your GitHub repository in your application or have a free-form PDF portfolio of your software development works as an attachment.

The working language is either Finnish or English. However, you should be able to read and write software documentation in English.

We offer a relaxed working environment and an inspiring view to computing education research. Feel free to ask further questions.

25. Ohjelmoinnin peruskurssi Y1: harjoitustehtävien suunnittelu 
Academic contact person for further information on topic: Kerttu Pollari-Malmi
Contact info: [email protected] 
Number of open positions: 2

Tehtävänä on keksiä uusia harjoitustehtäväideoita kurssille CS-A1111 Ohjelmoinnin peruskurssi Y1, kirjoittaa tehtävänannot suomeksi ja englanniksi sekä laatia tehtäville automaattiset tarkistimet A+-järjestelmään. Työssä vaaditaan ideointikykyä, hyvää ohjelmointitaitoa sekä hyvää suomen ja englannin kielen taitoa.

26. Physics-informed neural networks for data-driven discovery of the properties of turbulent transport in fluids 
Professor/Academic contact person for further information on topic: Maarit Korpi-Lagg/Ghassem Gozaliasl/Dale Weigt 
Contact info: [email protected], [email protected], [email protected]
Number of open positions: 1

Simulations of turbulent and magnetized fluids have recently advanced to more realistic parameter regimes thanks to their acceleration with graphics processing units (GPUs, see, e.g., [1]). Such simulations nowadays provide high-fidelity data of various interesting environments, where turbulence leads to non-trivial inverse cascade: large-scale flows and magnetic fields can arise driven by the chaotic motions. Some theoretical frameworks exist, providing the starting point for investigating, understanding, and modeling these phenomena. It is unclear, however, whether these frameworks are accurate enough to provide a working model for the turbulent transport. In this project we apply physics-informed neural networks [2] to investigate the validity of some of the standard frameworks, but also develop a new model with a data-driven discovery approach on the simulated data.

Prerequisites/Skills: Basics of deep learning (CS-E4890 Deep Learning), Programming (Python); experience with  CUDA/C++/C, Message-Passing Interface (MPI), and high-performance computing environments are a bonus.

[1] Pekkilä, J., Väisälä M. S., Korpi-Lagg, M. J., Rheinhardt, M., Lappi, O., Scalable communication for high-order stencil computations using CUDA-aware MPI, Parallel Computing, 111, 2022, 102904, https://doi.org/10.1016/j.parco.2022.102904.
[2] Cai, S., Mao, Z., Wang, Z. et al. Physics-informed neural networks (PINNs) for fluid mechanics: a review. Acta Mech. Sin. 37, 1727–1738 (2021)

27. Scalable Reinforcement Learning Systems for Training LLMs
Professor/Academic contact person for further information on topic: Bo Zhao
Contact info: [email protected]
Number of open positions: 2

Field of study: Deep Learning, Reinforcement Learning, Distributed Systems, Machine learning Systems, Big Data Systems

Machine learning (ML) systems translate data into value for decision making. Recent breakthroughs in large language models (GPT 4, Llama 2, ChatGPT, Gemini) and remarkable outcomes of reinforcement learning (RL) in real-world settings (AlphaGo, AlphaFold, reinforcement-learning-from-human-feedback) have shown that scalable model training on large GPU/TPU clusters is critical to obtain state-of-the-art performance.

This summer research project aims to answer the question “how to co-design multiple layers of the software/system stack to improve the scalability and performance of RL/RLHF systems”. Specifically, it addresses the challenges to build:
(1) flexible distributed RL systems to accelerate and parallelize the RL and RLHF training loop, 
(2) statement management libraries to transparently change the GPU device allocation and multi-dimensional parallelism (i.e., data/model/pipeline parallelism) without affecting the training result and 
(3) Holistic end-to-end training pipeline of RLHF training loop for LLMs.

The project aims at building open-source software and/or publishing at top-tier conferences.

About the lab. Aalto Data-Intensive System group (ADIS) conduct research on efficient data-intensive systems that translate data into value for decision making. The scope of our research spans across multiple subfields, from scalable data-centric machine learning systems to distributed data stream management systems, as well as code optimization techniques. Students have the chance to experience world-class research environment including Aalto HPC cluster Triton (200 GPUs) and the access to Europe's fastest (world’s 5th fastest) supercomputer LUMI equipped with quantum computing capacity. Students also have opportunities to collaborate with other world-leading research groups and industry labs within our international network (e.g., Imperial College, TUM, MPI-SWS, HU Berlin, NUS, Uni Edinburgh, AWS, etc).

Necessary skills. 
•    Solid knowledge of computer systems (e.g., distributed systems, data management systems, compilers)
•    Familiar with machine learning frameworks (e.g., PyTorch, TensorFlow, Megatron-LM, DeepSpeed)
•    Knowledge of distributed ML training (e.g., training across multiple GPUs or multiple nodes) 
•    Solid programming capability of C++, Python, Go and/or Rust, etc.
•    Strong analytical thinking skills
•    Excellent scientific communication and writing skills
•    Comprehensive interest in scientific problems and the ability to work independently and within a larger team
Preferred skills.
•    Hands-on experience of distributed ML training (e.g., model training across multiple GPUs or multiple nodes)

28. Stochastic algorithms for testing (un)knottedness
Professor/Academic contact person for further information on topic: Pekka Orponen
Contact info: [email protected]
Number of positions: 1

Mathematically speaking, a knot is an embedding of the 1-dimensional circle into 3-dimensional Euclidean space [1], and two knots are equivalent (or ambient isotopic) if they can be continuously deformed to each other. A knot is trivial or the unknot if it is equivalent to the round circle in 3-space. It is a notoriously difficult open question whether an efficient algorithm for testing (un)knottedness exists, although some remarkable advances have been achieved recently [2]. The task can nevertheless be approached in many ways that yield partial results, one of which is trying to find good sequences of Reidemeister moves [3] for unknotting a given knot diagram.
In the present context, the challenge of (un)knottedness arises in our work on the browser-based tool DNAforge [4] for designing 3D wireframe DNA nanostructures. Here the fundamental technique of DNA origami [5,6] builds on folding a long cyclical scaffold strand into an outline or “routing” of the desired target shape, guided by a large number of short staple strands. If the targeted strand routing is nontrivially knotted, then the physical folding cannot take place, and one should be aware of this in the design phase.
The task of this internship project is to adapt and test stochastic search techniques, such as Simulated Annealing [7], towards the goal of searching for a good Reidemeister sequence unknotting a given scaffold strand routing. If no such sequence can be found in a reasonable time, then the given routing is labelled as being at risk of being knotted. As part of the project, the selected method will be integrated as part of the DNAforge tool.
The project requires familiarity with basic algorithm design techniques, facility with combinatorial thinking, and good programming skills. Knowledge of biomolecules is not necessary, but familiarity with Javascript (or willingness to learn) is a prerequisite. For further information about our work, please see the research group webpage at https://research.cs.aalto.fi/nc/.
[1] https://en.wikipedia.org/wiki/Knot_(mathematics)
[2] https://www.maths.ox.ac.uk/node/38304
[3] https://en.wikipedia.org/wiki/Reidemeister_move
[4] https://dnaforge.org/
[5] https://doi.org/10.1038/nature04586
[6] https://doi.org/10.1038/nature14586
[7] https://en.wikipedia.org/wiki/Simulated_annealing

29. Research assistants for distributed systems 
Professor/Academic contact person for further information on topic: Senior Lecturer Vesa Hirvisalo
Contact info: [email protected]
Number of open positions: 2

We are looking for research assistants that could participate in our research on distributed systems. We do research on future Internet systems, for which we typically apply various analysis methods. In addition to programming work, our research also covers data engineering aspects.

To succeed in a research assistant position you should have a strong background on computer science and good programming skills. We are especially looking for B.Sc. level or M.Sc. level students that are close to making their related thesis works (optionally you can continue with the thesis work). However, other applicants will be also considered.

If you are interested, please contact us for discussing the options for your, or you can also directly send an application to us (if you do so, please add a CV and a study transcript in addition to a motivation letter).

30. Harjoitustehtävien ja niiden automaattitarkistimien laatiminen uudelle Tietokone työvälineenä -kurssille
Professor/Academic contact person for further information on topic: Jukka Suomela
Contact info: [email protected]  
Number of open positions: 2-3

The position requires a fluent command of written and spoken Finnish.

Tehtävänä on laatia yhdessä opetushenkilökunnan kanssa uusia harjoitustehtäviä sekä niiden automaattisia tarkistimia syksyllä 2024 alkavalle Tietokone työvälineenä -kurssille. Kurssin tehtävät koskevat Git-versionhallintaa, LaTeX-ladontajärjestelmää, komentorivityökalujen käyttöä sekä suurien tietomäärien käsittelyä, eli mitä voi tehdä, kun tiedosto on liian suuri Excelissä käsiteltäväksi. Tehtävä vaatii (ainakin suurimman osan) edellä mainittujen työkalujen hallintaa, ohjelmointitaitoa sekä hyvää ideointikykyä sekä yhteistyötaitoja ja aloitteellisuutta. Aiempi kokemus A+-järjestelmän automaattisten tarkistimien teosta on hyödyllinen.
Koska kurssin kieli on suomi, hyvä suomen kielen taito on välttämätön tässä työtehtävässä.

31. Y1-kurssin tentit ja Y2-kurssin tehtävät 
Professor/Academic contact person for further information on topic: Sanna Suoranta
Contact info: [email protected]
Number of open positions: 2-3

These positions require a fluent command of written Finnish. Behärskar svenska språket är en fördel. However, English is also needed.

Tässä tehtävässä tehdään tenttitehtäviä ja niiden automaattitarkistimia kurssille CS-A1111 Ohjelmoinnin peruskurssi Y1 sekä harjoitustehtäviä kurssille CS-A1121 Ohjelmoinnin peruskurssi Y2. Tehtävänantojen miettiminen vaatii luovuutta ja automaattitarkistimien tekeminen vaatii tarkkuutta ja kykyä ajatella monenlaisia ratkaisuja. Tehtävä vaatii ohjelmointitaitoa Python-kielellä. Aiempi kokekemus A+-järjestelmän automaattisten tarkistimien teosta on hyödyllinen. Kurssit ovat tarjolla suomeksi ja englanniksi (ja tentti myös ruotsiksi), joten vähintään suomea ja englantia tulee hallita erinomaisesti.

32. Security and dark patterns
Professor/Academic contact person for further information on topic: Sanna Suoranta
Contact info: [email protected]
Number of open positions: 1

Many webservices lure users to choose options that are best interest of the service, not the user. Such user interface techniques are called dark patterns. They may, for example, lead users to give permission to information they do not intended. We are making experiments and reviews how the dark patterns affect information security decisions made by common users. The work requires knowledge of user interface and user experience design, information security and programming.

33. Applying threat modelling tools for industrial automation scenarios 
Professor/Academic contact person for further information on topic: Mikko Kiviharju
Contact info: [email protected]
Number of open positions: 1

The applicant needs to learn / master the use of open-source threat modelling tools, such as Threagile or Threat Dragon, and be able to model an informal system description in YAML-language. Threat model is created based on IEC-62443 zone-conduit model (ZCR) of an example industrial automation system description, so familiarity with IEC-62443 ZCR is recommended.

34. Bayesian Workflow LLM Q&A
Professor/Academic contact person for further information on topic: Anna Riha 
Contact info: [email protected] 
Number of open positions: 1

A wide range of resources are available for learning about Bayesian data analysis and Bayesian modelling workflows, ranging from introductory and advanced textbooks and scientific papers to case studies and forum discussions. Especially for beginners, it is often not straightforward to identify where to find help fast and, despite available tools and resources, searching for tailored support on specific problems can be time-consuming. In this project, you will build a prototype of LLM-supported Q&A tool for Bayesian workflow.
Prerequisites: strong interest in Bayesian data analysis and Bayesian workflows familiarity with LLMs, chatGPT and similar tools; Langchain experience is a plus familiarity with Huggingface or similar platforms for deployment of interactive tools

35. Model Selection for Auto-Regressive Time-Series Models
Academic contact person for further information on topic: Dr. David Kohns 
Contact info: [email protected]
Number of open positions: 1

Model selection is an integral part of the Bayesian workflow. The projection predictive model selection framework in particular has shown promise to this end, and has become popular in many fields ranging from medical statistics to deep learning. Over the course of this project, we endeavour to extend projection predictive inference to auto-regressive time-series models using Bayesian Leave-Future-Out cross-validation. This works adds to the research-agenda started with McLatchie et al. (2022) (https://arxiv.org/abs/2208.14824). Students can expect their work to be an integral part of an academic publication and the code to be integrated in widely used Bayesian statistics packages in R. Pre-requisite: Coding experience in R, developing R packages, Bayesian statistics.

36. Bayesian workflow
Professor/Academic contact person for further information on topic: Aki Vehtari
Contact info: [email protected] 
Number of open positions: 1

You will take part in developing computational diagnostic tools for different parts of Bayesian workflow (see, e.g., https://arxiv.org/abs/2011.01808). Prerequisites: Bayesian inference and MCMC.

37. Theory of distributed and parallel computing 
Professor/Academic contact person for further information on topic: Jukka Suomela
Contact info: [email protected]
Number of open positions: 1

Our research group "Distributed Algorithms" is looking for a summer intern to help us with our research related to the theoretical foundations of distributed and parallel computing. We expect a good understanding of mathematics (especially in discrete math and graph theory) and algorithms and theoretical computer science. We also often try to outsource our work to computers, so if you have good programming skills and/or some knowledge of e.g. SAT solvers or proof assistants, it is a plus. We have also exciting opportunities for those who are interested in quantum computation in the distributed setting. For more information, see https://research.cs.aalto.fi/da/

38. Developers for the "Programming Parallel Computers" course
Professor/Academic contact person for further information on topic: Jukka Suomela
Contact info: [email protected]
Number of open positions: 1-2

We are hiring summer interns to help us with the development of the "Programming Parallel Computers" course. Some relevant keywords include C++, Rust, Python, Flask, SQL, HTML, CSS, Linux, Git, CI/CD, Ansible, and Docker (but you don't need to know everything, as long as you are prepared to learn). It is a plus if you have already taken the "Programming Parallel Computers" course; see https://ppc.cs.aalto.fi for more information.

39. Investigating Homophily and the Glass Ceiling in Supervisor- and Collaboration-Networks
Professor/Academic contact person for further information on topic: Barbara Keller
Contact info: [email protected]
Number of open positions: 1

Homophily and the Glass Ceiling are concepts from Sociology: "The glass ceiling is a colloquial term for the social barrier preventing women and members of minority groups from being promoted to top jobs in management" and "Homophily is describing the tendency of individuals to associate and bond with similar others". In "Homophily and the Glass Ceiling Effect in Social Networks" the authors described a graph evolution model which exhibits a glass ceiling effect under certain parameters. We want to extend this work by investigating additional real-world networks, such as (but not limited to), supervisor- and collaborator-networks.

The tasks involves:
- Finding relevant data sources
- Scraping and cleaning data
- Calculating relevant metrics
- Write-up of the findings
The applicant is interested in Social Networks and their analysis and has sound programming skills, preferably in python.

40. Intragroup and intergroup dynamics on Reddit
Professor/Academic contact person for further information on topic: Barbara Keller
Contact info: [email protected]
Number of open positions: 1

In this computational social science project we analyse real world social media. Specifically,  we study discussion dynamics within and between groups (e.g., fan groups, pro/anti groups of a certain topic) by analyzing Reddit data. This position is suitable for someone who has interest in the topic, solid coding skills (preferrably in Python), familiarity with the Reddit platform, and some knowledge of social media data analysis methods. Experience with analyzing social media data (especially Reddit data) will be a plus.

41. Summer internship positions in computer vision and machine learning
Professor/Academic contact person for further information on topic: Juho Kannala
Contact info: [email protected], +358503537744
Number of open positions: 3

Computer vision is a rapidly developing field that is at the forefront of recent advances in artificial intelligence. Our group has broad research interests within computer vision. We are pursuing problems both in geometric computer vision (including topics such as visual SLAM, visual-inertial odometry, optical flow, image-based 3D modeling and Gaussian splatting) and in semantic computer vision (including topics such as object detection and recognition, and deep learning). We are looking for students interested in both basic research and applications of computer vision. Students with good programming skills and strong background in mathematics are especially encouraged to apply. The precise topics of the research will be chosen together with the students to match their personal interests.

Examples of our recent papers include: https://aaltovision.github.io/PIVO/, https://aaltovision.github.io/pioneer/, https://aaltoml.github.io/GP-MVS/, https://github.com/AaltoVision/DGC-Net, https://github.com/AaltoVision/hscnet. For more papers and further information visit: https://users.aalto.fi/~kannalj1/

42. Designing for Developer Experience
Professor/Academic contact person for further information on topic: Fabian Fagerholm
Contact info: [email protected]

The Mind and Software research group is looking for skilled and motivated research assistants to contribute to our research on developer experience. Developer experience refers to the cognitive, motivational, and affective experience that software developers have while developing software. We design methods and tools for studying and assessing developer experience in various contexts in modern software development.

In this project, you will contribute to one or more ongoing effort related to the following themes:

• Software developers' mental models of modern programming frameworks.
• Developer experience of low-code development platforms.
• Longitudinal measurement of developer experience.

We are looking for people to perform different kinds of tasks, and you may be involved in one or more of them in different combinations. In research-oriented tasks, you participate in planning and conducting empirical studies with software developers, collect data through interviews, observation, or instrumentation, analyse data using qualitative and quantitative methods, and review existing scientific literature. In technically oriented tasks, you participate in developing software for developer experience measurement. Finally, in design-oriented tasks, you contribute user interface and visual designs for research materials, questionnaires, and measurement tool user interfaces.

Required skills (research-oriented):

• Familiarity with empirical studies (e.g., interviews, think-aloud, cognitive task analysis).
• Understanding of the basics of HCI and/or psychology (e.g., cognitive or social psychology).

Required skills (technically oriented):

• Familiarity with modern web development (e.g., HTML5, CSS, JavaScript, Python).
• Familiarity with mobile app development (e.g., Android, iOS, React Native).
• Familiarity with collaborative software development (e.g., Git, Continuous Integration).

Required skills (design-oriented):

• Familiarity with user interface design (particularly mobile and web).
• Ability to create visually appealing design elements following existing design guidelines and with an understanding of the research goals of the design.
• Familiarity with visual design tools (e.g., Figma, Miro).
• Understanding of the basics of HCI and/or psychology (e.g., cognitive or social psychology).

In addition, we particularly appreciate skills and/or interest in:

• Understanding of research instrument development (e.g., questionnaire design).
• Specific qualitative or quantitative research methods (e.g., thematic analysis, descriptive statistics, statistical analysis).
• Teamworking skills.
• Academic writing skills (in English).

The applicant is not required to be an expert in these areas but should display a good foundation and willingness to learn and develop their skills on their own initiative as well as in collaboration with other members of the research group.

43. Methods and models for Continuous Experimentation 
Professor/Academic contact person for further information on topic: Fabian Fagerholm
Contact info: [email protected]

The Mind and Software research group is looking for skilled and motivated research assistants to contribute to our research on Continuous Experimentation (CE). CE is an approach where field experiments with real users inform software product development, for example, through A/B testing. We are investigating methods and models for various aspects of the CE process and for different kinds of organisations and products.

In this project, you would contribute to ongoing research to support development of ways to identify and specify what to test in experiments, how to produce representative experiment objects to use in the experiments, and to understand how humans make decisions in the experimentation process. You would participate in conducting empirical studies with software practitioners, which may involve interviews, focus groups, or task-oriented studies and the analysis of data from these. The position can be combined with a Master's thesis if you are enrolled at Aalto University.

Required skills:

• An interest in software product development, both in general and using experiment-based methods in particular (e.g., A/B testing).
• Basic knowledge of experimental design.
• An understanding of research with human subjects.
• Ability to read and summarise scientific literature.
• Academic writing skills (in English).

In addition, we particularly appreciate skills and/or interest in some of the following:

• Interview methods (both individual and group) and analysis of interview data (e.g., thematic analysis, cognitive task analysis).
• Statistical modelling such as regression or multilevel modelling, or Bayesian networks.
• Creating prototypes of different fidelity for experiments (e.g., ranging from wireframes to working software prototypes).
• Understanding of the basics of HCI and/or psychology in human studies (e.g., cognitive or social psychology).

The applicant is not required to be an expert in these areas but should display a good foundation and willingness to learn and develop their skills on their own initiative as well as in collaboration with other members of the research group.

44. AR-assisted indoor navigation system development with Unity
Professor/Contact: Antti Ylä-Jääski
Contact info: [email protected] 
Number of open positions: 1

We have earlier developed an Indoor positioning and navigation system with Android and iOS. In this project, you would take the source code from Android/iOS and implement the same functionality with Unity. Some baseline code for Unity exists, however, it is to be discussed whether is it better to start from scratch or use some parts of the non-complete Unity SW basis.

Dong, J, Noreikis, M, XIAO, YU & Yla-Jaaski, A 2019, 'ViNav: A Vision-based Indoor Navigation System for Smartphones', IEEE Transactions on Mobile Computing, vol. 18, no. 6, pp. 1461-1475. https://doi.org/10.1109/TMC.2018.2857772

Noreikis, M, Xiao, Y & Ylä-Jääski, A 2017, SeeNav: Seamless and Energy-Efficient Indoor Navigation using Augmented Reality. in Proceedings of the Thematic Workshops of ACM Multimedia 2017: Thematic Workshops '17 . ACM, pp. 186-193, ACM Multimedia, Mountain View, United States, 23/10/2017. https://doi.org/10.1145/3126686.3126733