I gave a chat within the workshop on how the synthesis of logic and device Studying, In particular locations for instance statistical relational learning, can allow interpretability.
Keen on synthesizing the semantics of programming languages? We've got a new paper on that, approved at OOPSLA.
The Lab carries out study in synthetic intelligence, by unifying learning and logic, by using a recent emphasis on explainability
If you are attending NeurIPS this yr, it's possible you'll be interested in looking at our papers that touch on morality, causality, and interpretability. Preprints are available over the workshop web page.
Our paper (joint with Amelie Levray) on Finding out credal sum-merchandise networks has actually been approved to AKBC. These networks, coupled with other sorts of probabilistic circuits, are interesting mainly because they warranty that sure different types of chance estimation queries might be computed in time linear in the dimensions in the https://vaishakbelle.com/ network.
The posting, to appear from the Biochemist, surveys a number of the motivations and strategies for producing AI interpretable and dependable.
Thinking about coaching neural networks with logical constraints? Now we have a completely new paper that aims toward entire fulfillment of Boolean and linear arithmetic constraints on coaching at AAAI-2022. Congrats to Nick and Rafael!
Bjorn and I are promotion a 2 calendar year postdoc on integrating causality, reasoning and expertise graphs for misinformation detection. See right here.
We study arranging in relational Markov decision processes involving discrete and constant states and steps, and an not known quantity of objects (through probabilistic programming).
Along with colleagues from Edinburgh and Herriot Watt, Now we have set out the demand a new study agenda.
Paulius' work on algorithmic methods for randomly generating logic courses and probabilistic logic systems has actually been acknowledged towards the concepts and practise of constraint programming (CP2020).
The framework is applicable to a significant class of formalisms, which include probabilistic relational designs. The paper also scientific tests the synthesis issue in that context. Preprint below.
If you are attending AAAI this year, you might be interested in checking out our papers that touch on fairness, abstraction and generalized sum-product problems.
I gave a chat around the risks of synthetic intelligence and analysis priorities in the Intercontinental Growth Culture.