Jay van Zyl @ ecosystem.Ai

Jay van Zyl @ ecosystem.Ai

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.

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.

Improving life with IoT and Analytics

Last weekend I went to attend one event hosted by GE digital in association with T-HUB and Idea labs.

The event was about industrial IoT and Predix Platform developed by GE Digital for IIOT.

It was a very much informative event, and I also had the privilege to meet some awesome people over there who are part of this platform development.

GE is the well-known company we all know.   

Now coming to the usage of IoT and analytics in this. GE Develops a lot of turbines…

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