Gotta Learn Fast: A new benchmark for generalization in RL
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
#RSS原文
OpenAI News
OpenAI News Retro Contest
【RSS 原文 · 未经 AI 筛选或翻译】We’re launching a transfer learning contest that measures a reinforcement learning algorithm’s ability to generalize from previous experience.
OpenAI News Variance reduction for policy gradient with action-dependent factorized baselines
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Improving GANs using optimal transport
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Report from the OpenAI hackathon
【RSS 原文 · 未经 AI 筛选或翻译】On March 3rd, we hosted our first hackathon with 100 members of the artificial intelligence community.
OpenAI News On first-order meta-learning algorithms
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Reptile: A scalable meta-learning algorithm
【RSS 原文 · 未经 AI 筛选或翻译】We’ve developed a simple meta-learning algorithm called Reptile which works by repeatedly sampling a task, performing stochastic gradient descent on it, and updating the initial parameters towards the final parameters learned on that task. Reptile is the application of the Shortest Descent algorithm to the meta-learning setting, and is mathematically similar to first-order MAML (which is a version of the well-known MAML algorithm) that only needs black-box access to an optimizer such as SGD or Adam, with similar computational efficiency and performance.
OpenAI News OpenAI Scholars
【RSS 原文 · 未经 AI 筛选或翻译】We’re providing 6–10 stipends and mentorship to individuals from underrepresented groups to study deep learning full-time for 3 months and open-source a project.
OpenAI News Some considerations on learning to explore via meta-reinforcement learning
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Ingredients for robotics research
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing eight simulated robotics environments and a Baselines implementation of Hindsight Experience Replay, all developed for our research over the past year. We’ve used these environments to train models which work on physical robots. We’re also releasing a set of requests for robotics research.
OpenAI News Multi-Goal Reinforcement Learning: Challenging robotics environments and request for research
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News OpenAI hackathon
【RSS 原文 · 未经 AI 筛选或翻译】Come to OpenAI’s office in San Francisco’s Mission District for talks and a hackathon on Saturday, March 3rd.
OpenAI News Preparing for malicious uses of AI
【RSS 原文 · 未经 AI 筛选或翻译】We’ve co-authored a paper that forecasts how malicious actors could misuse AI technology, and potential ways we can prevent and mitigate these threats. This paper is the outcome of almost a year of sustained work with our colleagues at the Future of Humanity Institute, the Centre for the Study of Existential Risk, the Center for a New American Security, the Electronic Frontier Foundation, and others.
OpenAI News OpenAI supporters
【RSS 原文 · 未经 AI 筛选或翻译】We’re excited to welcome new donors to OpenAI.
OpenAI News Interpretable machine learning through teaching
【RSS 原文 · 未经 AI 筛选或翻译】We’ve designed a method that encourages AIs to teach each other with examples that also make sense to humans. Our approach automatically selects the most informative examples to teach a concept—for instance, the best images to describe the concept of dogs—and experimentally we found our approach to be effective at teaching both AIs
OpenAI News Discovering types for entity disambiguation
【RSS 原文 · 未经 AI 筛选或翻译】We’ve built a system for automatically figuring out which object is meant by a word by having a neural network decide if the word belongs to each of about 100 automatically-discovered “types” (non-exclusive categories).
OpenAI News Requests for Research 2.0
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing a new batch of seven unsolved problems which have come up in the course of our research at OpenAI.
OpenAI News Scaling Kubernetes to 2,500 nodes
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Block-sparse GPU kernels
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the chosen sparsity, these kernels can run orders of magnitude faster than cuBLAS or cuSPARSE. We’ve used them to attain state-of-the-art results in text sentiment analysis and generative modeling of text and images.
OpenAI News Learning sparse neural networks through L₀ regularization
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Interpretable and pedagogical examples
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Learning a hierarchy
【RSS 原文 · 未经 AI 筛选或翻译】We’ve developed a hierarchical reinforcement learning algorithm that learns high-level actions useful for solving a range of tasks, allowing fast solving of tasks requiring thousands of timesteps. Our algorithm, when applied to a set of navigation problems, discovers a set of high-level actions for walking and crawling in different directions, which enables the agent to master new navigation tasks quickly.
OpenAI News Generalizing from simulation
【RSS 原文 · 未经 AI 筛选或翻译】Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the environment as they solve simple tasks. That is, we’ve used these techniques to build closed-loop systems rather than open-loop ones as before.
OpenAI News Asymmetric actor critic for image-based robot learning
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Sim-to-real transfer of robotic control with dynamics randomization
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Domain randomization and generative models for robotic grasping
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Meta-learning for wrestling
【RSS 原文 · 未经 AI 筛选或翻译】We show that for the task of simulated robot wrestling, a meta-learning agent can learn to quickly defeat a stronger non-meta-learning agent, and also show that the meta-learning agent can adapt to physical malfunction.
OpenAI News Competitive self-play
【RSS 原文 · 未经 AI 筛选或翻译】We’ve found that self-play allows simulated AIs to discover physical skills like tackling, ducking, faking, kicking, catching, and diving for the ball, without explicitly designing an environment with these skills in mind. Self-play ensures that the environment is always the right difficulty for an AI to improve. Taken alongside our Dota 2 self-play results, we have increasing confidence that self-play will be a core part of powerful AI systems in the future.
OpenAI News Nonlinear computation in deep linear networks
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Learning to model other minds
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing an algorithm which accounts for the fact that other agents are learning too, and discovers self-interested yet collaborative strategies like tit-for-tat in the iterated prisoner’s dilemma. This algorithm, Learning with Opponent-Learning Awareness (LOLA), is a small step towards agents that model other minds.
OpenAI News Learning with opponent-learning awareness
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News OpenAI Baselines: ACKTR & A2C
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing two new OpenAI Baselines implementations: ACKTR and A2C. A2C is a synchronous, deterministic variant of Asynchronous Advantage Actor Critic (A3C) which we’ve found gives equal performance. ACKTR is a more sample-efficient reinforcement learning algorithm than TRPO and A2C, and requires only slightly more computation than A2C per update.
OpenAI News More on Dota 2
【RSS 原文 · 未经 AI 筛选或翻译】Our Dota 2 result shows that self-play can catapult the performance of machine learning systems from far below human level to superhuman, given sufficient compute. In the span of a month, our system went from barely matching a high-ranked player to beating the top pros and has continued to improve since then. Supervised deep learning systems can only be as good as their training datasets, but in self-play systems, the available data improves automatically as the agent gets better.
OpenAI News Dota 2
【RSS 原文 · 未经 AI 筛选或翻译】We’ve created a bot which beats the world’s top professionals at 1v1 matches of Dota 2 under standard tournament rules. The bot learned the game from scratch by self-play, and does not use imitation learning or tree search. This is a step towards building AI systems which accomplish well-defined goals in messy, complicated situations involving real humans.
OpenAI News Gathering human feedback
【RSS 原文 · 未经 AI 筛选或翻译】RL-Teacher is an open-source implementation of our interface to train AIs via occasional human feedback rather than hand-crafted reward functions. The underlying technique was developed as a step towards safe AI systems, but also applies to reinforcement learning problems with rewards that are hard to specify.
OpenAI News Better exploration with parameter noise
【RSS 原文 · 未经 AI 筛选或翻译】We’ve found that adding adaptive noise to the parameters of reinforcement learning algorithms frequently boosts performance. This exploration method is simple to implement and very rarely decreases performance, so it’s worth trying on any problem.
OpenAI News Proximal Policy Optimization
【RSS 原文 · 未经 AI 筛选或翻译】We’re releasing a new class of reinforcement learning algorithms, Proximal Policy Optimization (PPO), which perform comparably or better than state-of-the-art approaches while being much simpler to implement and tune. PPO has become the default reinforcement learning algorithm at OpenAI because of its ease of use and good performance.
OpenAI News Robust adversarial inputs
【RSS 原文 · 未经 AI 筛选或翻译】We’ve created images that reliably fool neural network classifiers when viewed from varied scales and perspectives. This challenges a claim from last week that self-driving cars would be hard to trick maliciously since they capture images from multiple scales, angles, perspectives, and the like.
OpenAI News Hindsight Experience Replay
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。
OpenAI News Teacher–student curriculum learning
【RSS 原文 · 未经 AI 筛选或翻译】此订阅未提供摘要，请通过原文链接阅读。