Since its beta release in 2017, DVC has become an essential tool for many data science teams. Its data versioning capabilities, reproducible pipelines, and experiment tracking features are at the core of our ecosystem of open MLOps tools. Today we are proud to launch a new product that extends how machine learning teams can use DVC: our extension for Visual Studio Code. With this extension, you get a full VS Code-native experimentation platform for your machine learning projects. Control your datasets and models, run experiments, view metrics, create plots, and much more.

All living organisms carve out environmental niches within which they can maintain relative predictability amidst the ever-increasing entropy around them (1), (2). Humans, for example, go to great lengths to shield themselves from surprise -- we band together in millions to build cities with homes, supplying water, food, gas, and electricity to control the deterioration of our bodies and living spaces amidst heat and cold, wind and storm. The need to discover and maintain such surprise-free equilibria has driven great resourcefulness and skill in organisms across very diverse natural habitats. Motivated by this, we ask: could the motive of preserving order amidst chaos guide the automatic acquisition of useful behaviors in artificial agents? This central problem in artificial intelligence has evoked several candidate solutions, largely focusing on novelty-seeking behaviors (3), (4), (5).

This is frame-by-frame prediction with no temporal constraints. This is a frame-by-frame keypoint prediction on each animation frame. No temporal information is used. We show how the network is able to utilize the same keypoints across object instances and consistently predict keypoints across viewing angles, even when parts are occluded such as the back legs. Your browser does not support the video tag.

Abstract: Deep learning algorithms have recently appeared that pretrain hidden layers of neural networks in unsupervised ways, leading to state-of-the-art performance on large classification problems. These methods can also pretrain networks used for reinforcement learning. However, this ignores the additional information that exists in a reinforcement learning paradigm via the ongoing sequence of state, action, new state tuples. This paper demonstrates that learning a predictive model of state dynamics can result in a pretrained hidden layer structure that reduces the time needed to solve reinforcement learning problems. After training for 0 minutes: Your browser does not support the video tag.