DLPOTW: DISCOVERING SYMBOLIC MODELS FROM DEEP LEARNING WITH INDUCTIVE BIASES

🚀 DEEP LEARNING PAPER OF THE WEEK Miles Cranmer, Alvaro Sanchez-Gonzalez, Peter Battaglia, Rui Xu, Kyle Cranmer, David Spergel, Shirley Ho Abstract "We develop a general approach to distill symbolic representations of a learned deep model by introducing strong inductive biases. We focus on Graph Neural Networks (GNNs). The technique works as follows: we first encourage sparse latent representations when we train … Continue reading DLPOTW: DISCOVERING SYMBOLIC MODELS FROM DEEP LEARNING WITH INDUCTIVE BIASES

PAPERS WITH ANNOTATIONS

MLT Co-Director Alisher Abdulkhaev shares his Papers with Annotations. This project compiles multiple (AI related) papers with illustrations, annotations, and brief explanations of technical keywords, terms and previous studies which makes it easy to read the paper and follow the main idea. Object detection papers with annotationsAI and Cognitive Science related papers with annotations Please … Continue reading PAPERS WITH ANNOTATIONS

MLT WOMEN IN MACHINE LEARNING

More than 100 people joined us for our 4th MLT Women in Machine Learning, this time it was held at the brand new Google Japan office in the impressive Shibuya Stream building. We heard two fantastic talks – one on production-level ML, the other from a research perspective – and had a great panel discussion … Continue reading MLT WOMEN IN MACHINE LEARNING

MLT AT NEURIPS 2019 IN VANCOUVER

MLT is presenting 2 workshop papers at the NeurIPS 2019 conference in Vancouver, Canada. Meet Asir Saeed at the poster session at the NeurIPS Workshop Machine Learning for Creativity and Design. Creative GANs for generating poems, lyrics, and metaphors Asir Saeed, Suzana Ilić, Eva Zangerle Generative models for text have substantially contributed to tasks like machine translation … Continue reading MLT AT NEURIPS 2019 IN VANCOUVER