Accelerating reinforcement learning through GPU Atari emulation
Politecnico di Milano
this event will be online and organized by Microsoft Teams
April 22nd, 2020
5.30 pm
Contacts:
Elena De Momi
this event will be online and organized by Microsoft Teams
April 22nd, 2020
5.30 pm
Contacts:
Elena De Momi
Sommario
We introduce CuLE (CUDA Learning Environment), a CUDA port of the Atari Learning Environment (ALE) which is used for the development of deep reinforcement algorithms. CuLE overcomes many limitations of existing CPU-based emulators and scales naturally to multiple GPUs. It leverages GPU parallelization to run thousands of games simultaneously and it renders frames directly on the GPU, to avoid the bottleneck arising from the limited CPU-GPU communication bandwidth. CuLE generates up to 155M frames per hour on a single GPU, a finding previously achieved only through a cluster of CPUs. Beyond highlighting the differences between CPU and GPU emulators in the context of reinforcement learning, we show how to leverage the high throughput of CuLE by effective batching of the training data, and show accelerated convergence for A2C+V-trace. CuLE is available at https://github.com/NVlabs/cule.
This event will be online and organized via Microsoft Teams.
In order to attend the seminar, please, go to the following link: Join Microsoft Teams Meeting.
This event will be online and organized via Microsoft Teams.
In order to attend the seminar, please, go to the following link: Join Microsoft Teams Meeting.
Biografia
Iuri Frosio got his PhD in biomedical engineering at the Politecnico of Milan in 2006. He was a research fellow at the Computer Science Department of the University of Milan from 2003 and an assistant professor in the same department from 2006 to 2013. In the same period, he worked as a consultant for various companies in Italy and in the US. He joined NVIDIA in 2014 as senior research scientist. His research interests include image processing, computer vision, robotics, parallel programming, machine learning, and reinforcement learning.