Engineering Insights

Node.js 6.11.3 (LTS)

OpenAI Baselines: ACKTR & A2C

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.

More on Dota 2

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.

Node.js 8.4.0 (Current)

Dota 2

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.

Node.js 8.3.0 (Current)

Next.js 3.0

We are very excited excited to announce the stable release of Next.js 3.0. Ever since our , we have been using it to power  and have received lots of feedback and contributions from our .beta announcementvercel.comcommunity Let’s walk through what’s been improved and what’s altogether new, or fetch the latest version from !npm  Next.js is a zero-configuration, single-command toolchain for React apps, with built-in server-rendering, code-splitting and more. Check out  to get started!New to Next.j

Gathering human feedback

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.