Grounding Neural Inference with Satisfiability Modulo Theories -Leveraging Theory Solvers for Machine Learning
Dec 10, 2023
Speakers
Saranya Vijayakumar
Speaker · 0 followers
Zifan Wang
Speaker · 0 followers
Vijay Ganesh
Speaker · 0 followers
Somesh Jha
Speaker · 1 follower
Matt Fredrikson
Speaker · 0 followers
About
Recent techniques that integrate solver layers into Deep Neural Networks (DNNs) have shown promise in bridging a long-standing gap between inductive learning and symbolic reasoning techniques. In this paper we present a set of techniques for integrating Satisfiability Modulo Theories (SMT) solvers into the forward and backward passes of a deep network layer, called SMTLayer.Using this approach, one can encode rich domain knowledge into the network in the form of mathematical formulas.In the forward pass, the solver uses symbols produced by prior layers, along with these formulas, to construct inferences; in the backward pass, the solver informs updates to the network, driving it towards representations that are compatible with the solver's theory.Notably, the solver need not be differentiable. We implement SMTLayer as a Pytorch module, and our empirical results show that it leads to models that 1) require fewer training samples than conventional models, 2) that are robust to certain types of covariate shift, and 3) that ultimately learn representations that are consistent with symbolic knowledge, and thus naturally interpretable.Recent techniques that integrate solver layers into Deep Neural Networks (DNNs) have shown promise in bridging a long-standing gap between inductive learning and symbolic reasoning techniques. In this paper we present a set of techniques for integrating Satisfiability Modulo Theories (SMT) solvers into the forward and backward passes of a deep network layer, called SMTLayer.Using this approach, one can encode rich domain knowledge into the network in the form of mathematical formulas.In the forw…
Organizer
NeurIPS 2023
Account · 645 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Formulating Discrete Probability Flow Through Optimal Transport
Watch later
Pengze Zhang, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Amorphous Fortress: Exploring Emergent Behavior in Open-Ended Simulations
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
State-Action Similarity-Based Representations for Off-Policy Evaluation
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Multimodal Decoding of Human Brain Activity into Images and Text
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Robust Learning with Progressive Data Expansion Against Spurious Correlation
Watch later
Yihe Deng, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
SLaM: Student Label Mixing for Distillation with Unlabeled Examples
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%