Published inTDS ArchiveNeuro-algorithmic Policies: Can Combinatorics Help Reinforcement Learning?In this blog post we explore an application of blackbox differentiation to the setting of imitation learning, which leads to considerable…Apr 11, 2022A response icon2Apr 11, 2022A response icon2
Published inTDS ArchiveUnderstanding Monte Carlo EstimationMonte Carlo estimation is an essential part of a machine learning engineers’ toolbox.May 2, 2021May 2, 2021
Published inTDS ArchiveFundamental Problems of Probabilistic InferenceWhy should you care about sampling if you are a machine learning practitioner?Aug 26, 2020Aug 26, 2020
Published inTDS ArchiveNo Stress Gaussian ProcessesHow do you deal with a distribution over an infinite number of functions?Aug 23, 2020A response icon1Aug 23, 2020A response icon1
Published inTDS ArchiveThe Game of Life, the Legacy of John ConwayWhat is the Game of Life? What is the significance of the Game of Life? The legacy of the deceased John Conway.Apr 14, 2020A response icon1Apr 14, 2020A response icon1
Published inResearchers’ DigestFinding Causal Models is HardWhy is it so hard to find Structural Causal Models? A DAG perspective.Apr 8, 2020A response icon2Apr 8, 2020A response icon2
Published inTDS ArchiveControl What You Can: Reinforcement Learning with Task Planning!Here I talk about our NeurIPS 2019 paper, combining planning with reinforcement learning agents and intrinsic motivation.Apr 8, 2020Apr 8, 2020
Published inTDS ArchiveRaMBO: Ranking Metric Blackbox OptimizationOur paper resulting in an oral at CVPR 2020 about applying the blackbox differentiation theory (codename #blackboxbackprop) to optimizing…Apr 8, 2020Apr 8, 2020
Published inTDS Archive7 Things to Think About When Developing Reinforcement LearnersAlbeit we have made very good progress in reinforcement learning research, a unified framework to compare the algorithms is missing…Mar 23, 2020Mar 23, 2020
Published inTDS ArchiveWhat is the “Information” in Information Theory?Breaking down the fundamental concept of information.Feb 27, 2020A response icon2Feb 27, 2020A response icon2