Gods and Robots In this episode of the podcast we shake things up! Neil is on the guest side of the table with his partner Rabbi Laura Janner-Klausner to discuss their upcoming project Gods and Robots. Katherine is joined on the host side by friend of the show professor Michael Littman. See... See More Episodes arXiv Whitepapers Progress measures for grokking via mechanistic interpretability Neural networks often exhibit emergent behavior, where qualitatively new capabilities arise from scaling up the amount of parameters, training data, or training steps. One approach to understanding emergence is to find continuous \textit{progress measures} that underlie the seemingly discontinuous... Relative representations enable zero-shot latent space communication Neural networks embed the geometric structure of a data manifold lying in a high-dimensional space into latent representations. Ideally, the distribution of the data points in the latent space should depend only on the task, the data, the loss, and other architecture-specific constraints. However... Building Scalable Video Understanding Benchmarks through Sports Existing benchmarks for evaluating long video understanding falls short on multiple aspects, either lacking in scale or quality of annotations. These limitations arise from the difficulty in collecting dense annotations for long videos (e.g. actions, dialogues, etc.), which are often obtained by... More featured content News Articles Wireless movement-tracking system could collect health and behavioral data Merging cell datasets, panorama style Stay in the loop. Subscribe to our newsletter for a weekly update on the latest podcast, news, events, and jobs postings. E-mail Leave this field blank Citation Statistics of Machine Learning Papers Canadian health care needs agile leaders and bold visions for the future Smarter training of neural networks Using AI to predict breast cancer and personalize care Putting vision models to the test Breast cancer diagnosis by AI now as good as human experts New approach could accelerate efforts to catalogue vast numbers of cells For better deep neural network vision, just add feedback (loops) More news