Project description
Automatic differentiation systems in probabilistic programmes
Automatic differentiation (AD) systems like TensorFlow and probabilistic programming languages (PPLs) such as Stan simplify complex calculations in machine learning. However, their effectiveness is limited by a lack of theoretical foundations for composable programming, especially for applying AD to probabilistic choices and integrating Bayesian inference algorithms. The ERC-funded FoRECAST project aims to advance programming language theory and tools to enable efficient computations with derivatives and probabilities. The project will tackle real-world modelling challenges by collaborating with domain experts on case studies while establishing semantic foundations and algorithms for composable AD in probabilistic programs. It will develop a practical stochastic AD system that incorporates innovative gradient estimation techniques and a user-friendly PPL for composable Bayesian inference.
Objective
Automatic Differentiation (AD) systems, like TensorFlow, and probabilistic programming languages (PPLs), like Stan, automate complex computations of derivatives and Bayesian inference tasks. By stream- lining these computations for non-expert users, these high-level systems have accelerated progress across science and society (e.g. by enabling machine learning). Yet, the theoretical foundations needed to build a high-level system for composable programming with derivatives and probabilities are missing. This chasm in our knowledge severely limits the implementation of machine learning techniques, preventing them from reaching their full potential. Specifically, we do not understand (a) how to perform AD on programs built using probabilistic choices and expected values or (b) how to compose (i.e. combine and integrate) Bayesian inference algorithms. FoRECAST addresses this chasm by developing programming language theory and tools for flexible, composable, and efficient calculations with derivatives and prob- abilities. WP 1 develops case studies in collaboration with domain experts, to ensure that FoRECAST creates theory and systems relevant to real-world, complex modelling problems. WP 2 develops the semantic foundations, algorithms, and formalised correctness proofs for composable AD of probabilistic programs. WP 3 builds a practical stochastic (i.e. probabilistic) AD system that synthesises these novel gradient estimation techniques. WP 4 establishes theoretical foundations to compose Bayesian inference algorithms in PPLs. WP 5 implements a user-friendly PPL that facilitates composable Bayesian inference, enabling more flexible modelling for a wider user base. By mathematically formalising, generalising, optimising, and implementing a next generation PPL, this project will lay a trustworthy foundation upon which probabilistic data analysis applications (e.g. reinforcement learning, proteomics modelling, and paleoclimate reconstructions) can rise to the next level.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencesbiological sciencesbiochemistrybiomoleculesproteinsproteomics
- natural sciencescomputer and information sciencessoftwaresoftware applicationssimulation software
You need to log in or register to use this function
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Topic(s)
Funding Scheme
HORIZON-ERC - HORIZON ERC GrantsHost institution
3584 CS Utrecht
Netherlands