romerocesar

3 ms·

1. 1.▲
   by romerocesar 9y ago

   this enables RL research that supports multi-agent environments, multiple views (cameras) and doesn't need to pick between physics or graphics because a game engine can do both! this is cool
2. 2.▲
   by romerocesar 10y ago

   The world is amazing
3. 3.▲
   by romerocesar 10y ago

   srsly? all the discussion above 10hrs+ before your comment and that's your question? RTFM: https://github.com/amznlabs/amazon-dsstne/blob/master/FAQ.md
4. 4.▲
   by romerocesar 10y ago

   One important difference is model-parallel training. From the FAQ: DSSTNE instead uses “model-parallel training”, where each layer of the network is split across the available GPUs so each operation just runs faster. Model-parallel training