I gave a talk for the workshop on how the synthesis of logic and machine Studying, especially regions including statistical relational Discovering, can permit interpretability.
Final 7 days, I gave a talk in the pint of science on automatic units as well as their impression, touching on the topics of fairness and blameworthiness.
The Lab carries out exploration in synthetic intelligence, by unifying Studying and logic, by using a the latest emphasis on explainability
I attended the SML workshop in the Black Forest, and talked about the connections among explainable AI and statistical relational Mastering.
Gave a talk this Monday in Edinburgh around the concepts & exercise of machine Finding out, covering motivations & insights from our study paper. Crucial concerns lifted provided, the best way to: extract intelligible explanations + modify the product to fit changing requires.
A consortia undertaking on dependable units and goverance was accepted late final 12 months. Information connection here.
Serious about teaching neural networks with rational constraints? We have a completely new paper that aims toward total satisfaction of Boolean and linear arithmetic constraints on schooling at AAAI-2022. Congrats to Nick and Rafael!
The posting introduces a standard logical framework for reasoning about discrete and constant probabilistic models in dynamical domains.
Hyperlink In the final 7 days of October, I gave a talk informally speaking about explainability and moral duty in synthetic intelligence. Due to the organizers for your invitation.
Jonathan’s paper considers a lifted approached to weighted model integration, including circuit construction. Paulius’ paper develops a measure-theoretic perspective on weighted model counting and proposes a means to encode conditional weights on literals analogously to conditional probabilities, which results in significant overall performance improvements.
Extended abstracts of our NeurIPS paper (on PAC-Studying in to start with-purchase logic) and the journal paper on abstracting probabilistic styles was accepted to KR's a short while ago revealed investigation track.
A journal paper on abstracting probabilistic designs has actually been approved. The paper research the semantic constraints which allows one particular to abstract a fancy, low-degree product with a less complicated, superior-degree just one.
The initial introduces a primary-order language for reasoning about probabilities in dynamical domains, and the 2nd considers the automated fixing of chance troubles laid out in organic https://vaishakbelle.com/ language.
Our get the job done (with Giannis) surveying and distilling strategies to explainability in equipment Discovering has been recognized. Preprint here, but the ultimate Variation are going to be on the web and open up entry quickly.