A new way to do controlled experiments in medicine: simulate the control
Generative AI could augment randomized controlled trials.
Aalto computer scientists in AISTATS 2024
This year two papers from Aalto Department of Computer Science were accepted to the AISTATS conference
The 27th International Conference on Artificial Intelligence and Statistics (AISTATS) will be held on May 2nd - May 4th, 2024 in Valencia, Spain. AISTATS is an interdisciplinary gathering of researchers at the intersection of computer science, artificial intelligence, machine learning, statistics, and related areas.
Accepted papers
In alphabetical order. Click the title to see the authors and the abstract.
Authors
Manuel Haussmann, Tran Minh Son Le, Viivi Halla-aho, Samu Kurki, Jussi Leinonen, Miika Koskinen, Samuel Kaski, and Harri Lähdesmäki
Abstract
Randomized controlled trials (RCTs) are the accepted standard for treatment effect estimation but they can be infeasible due to ethical reasons and prohibitive costs. Single-arm trials, where all patients belong to the treatment group, can be a viable alternative but require access to an external control group. We propose an identifiable deep latent-variable model for this scenario that can also account for missing covariate observations by modeling their structured missingness patterns. Our method uses amortized variational inference to learn both group-specific and identifiable shared latent representations, which can subsequently be used for (i) patient matching if treatment outcomes are not available for the treatment group, or for (ii) direct treatment effect estimation assuming outcomes are available for both groups. We evaluate the model on a public benchmark as well as on a data set consisting of a published RCT study and real-world electronic health records. Compared to previous methods, our results show improved performance both for direct treatment effect estimation as well as for effect estimation via patient matching.
Authors
Prakhar Verma, Vincent Adam, and Arno Solin
Abstract
Diffusion processes are a class of stochastic differential equations (SDEs) providing a rich family of expressive models that arise naturally in dynamic modelling tasks. Probabilistic inference and learning under generative models with latent processes endowed with a non-linear diffusion process prior are intractable problems. We build upon work within variational inference approximating the posterior process as a linear diffusion process and point out pathologies in the approach. We propose an alternative parameterization of the Gaussian variational process using a site-based exponential family description. This allows us to trade a slow inference algorithm with fixed-point iterations for a fast algorithm for convex optimization akin to natural gradient descent, which also provides a better objective for the learning of model parameters
Department of Computer Science
cs.aalto.fi
School of Science
Science for tomorrow’s technology, innovations and businesses
- Published:
- Updated:
Read more news
Civil Engineering invites companies to participate in future development work
A stakeholder event organized by the Department of Civil Engineering has already become a spring tradition. Civil Engineering Day was held on April 25 in Dipoli Otaniemi.
Aalto Open Science Award ceremony brought together Aaltonians to discuss open science
Last week we gathered at A Grid to celebrate the awardees of the Aalto Open Science Award 2023 and discuss open science matters with the Aalto community.
Seed funding available to boost collaboration between Aalto, KU Leuven and University of Helsinki
Aalto University, KU Leuven and the University of Helsinki launch the 2nd exploratory seed funding call to explore research collaboration possibilities. The funding call is open until 10 September 2024.