Neural information processing systems have benefited tremendously from the availability of programming languages and frameworks for automatic differentiation (AD). Not only do NeurIPS benefit from programming languages for automatic inference but can also be considered as a language in their own right, consisting of differentiable and stochastic primitives. Combined with neural language models, these systems are increasingly capable of generating symbolic programs a human programmer might write in a high-level language. Developing neurosymbolic systems for automatic program synthesis requires insights from both statistical learning and programming languages.
AIPLANS invites all researchers working towards the same purpose in these two communities to build on common ground. Our workshop is designed to be as inclusive as possible towards researchers engaged in building programming languages and neurosymbolic systems.
| Introductory remarks (Introductory Remarks) | |
| Thinking like Transformers - Gail Weiss - Technion - Israel Institute of Technology (Invited Talk) | |
| Q&A - Gail Weiss (Post Talk Q & A) | |
| When Gödel discovered Automatic Differentiation - Marie Kerjean - Centre national de la recherche scientifique (Invited Talk) | |
| Building machines that learn and think like people by learning to write programs: progress, open problems, and next steps - Josh Tenenbaum - Massachusetts Institute of Technology (Invited Talk) | |
| Short break (Break) | |
| Panel Discussion | |
| Daniel Selsam Microsoft Research (Tutorial) | |
| Q&A - Daniel Selsam (Post Talk Q & A) | |
| Lunch / Poster Session (Poster Session) | |
| Remarks from Organisers (Introduction) | |
| Randomized Automatic Differentiation - Ryan Adams - Princeton University (Invited Talk) | |
| Q&A - Ryan Adams (Post Talk Q & A) | |
| Dependent Types for Machine Learning in Dex - David Duvenaud - University of Toronto (Invited Talk) | |
| Q&A - David Duvenaud (Post Talk Q & A) | |
| Differential Inference: A Criminally Underused Tool. - Alexander Rush - Cornell University (Invited Talk) | |
| Q&A - Alexander Rush (Post Talk Q & A) | |
| Introduction to Spotlight Speakers (Organiser Remarks) | |
| Meta-Learning an Inference Algorithm for Probabilistic Programs - Gwonsoo Che (Spotlight Talks) | |
| LazyPPL: laziness and types in non-parametric probabilistic programs - Hugo Paquet (Spotlight Talk) | |
| Learning Rules with Stratified Negation in Differentiable ILP - Giri Krishnan (Spotlight Talks) | |
| Learning Adaptive Control Flow in Transformers for Improved Systematic Generalization - Róbert Csordás (Spotlight Talk) | |
| Type Inference as Optimization - Eirene V. Pandi (Spotlight Talk) | |
| Q&A for Spotlight Authors (Q & A) | |
| Closing Remarks (Closing remarks) | |
| Poster Session | |
| A Genetic Programming Approach To Zero-Shot Neural Architecture Ranking (Poster) | |
| Meta-Learning an Inference Algorithm for Probabilistic Programs (Poster) | |
| Augmenting Classic Algorithms with Neural Components for Strong Generalisation on Ambiguous and High-Dimensional Data (Poster) | |
| Synthesizing Video Trajectory Queries (Poster) | |
| Learning C to x86 Translation: An Experiment in Neural Compilation (Poster) | |
| Adversarial Robustness of Program Synthesis Models (Poster) | |
| Learning Adaptive Control Flow in Transformers for Improved Systematic Generalization (Poster) | |
| Learning compositional programs with arguments and sampling (Poster) | |
| PAC Synthesis of Machine Learning Programs (Poster) | |
| Learning Rules with Stratified Negation in Differentiable ILP. (Poster) | |
| AutoCoder: Leveraging Transformers for Automatic Code Synthesis (Poster) | |
| Safe Neurosymbolic Learning with Differentiable Symbolic Execution (Poster) | |
| Staged compilation of tensor expressions (Poster) | |
| Towards Neural Functional Program Evaluation (Poster) | |
| Are Transformers All That Karel Needs? (Poster) | |
| Proof Extraction for Logical Neural Networks (Poster) | |
| AutumnSynth: Synthesis of Reactive Programs with Structured Latent State (Poster) | |
| LazyPPL: laziness and types in non-parametric probabilistic programs (Poster) | |
| Type Inference as Optimization (Poster) | |
| Scallop: From Probabilistic Deductive Databases to Scalable Differentiable Reasoning (Poster) | |