Solving Atari games with Distributed Reinforcement Learning

Solving #Atari games with distributed reinforcement learning. #MachineLearning

  • One particular step towards achieving this ideal was distributing a state-of-the-art Reinforcement Learning algorithm on a large CPU cluster, allowing super-fast training of agents that learned to master a wide range of Atari 2600 games.
  • The A3C algorithm developed later was able to achieve human performance in most games and did so with a similar amount of training time.
  • By distributing the BA3C (details of single-machine implementation here) reinforcement learning algorithm, we were able to make an agent teach itself to play a wide range of Atari games rapidly, by just looking at a raw pixel output (game screen) from the game emulator.
  • In the game of Breakout, our agent achieves a superhuman score in just 20 minutes, which is a significant reduction of the single machine implementation learning time.
  • The development of a distributed evolution strategy (ES) algorithm has led researchers from OpenAI to train agents to play Atari games in one hour by using 720 parallel CPUs.

We present the result of research conducted at deepsense.ai, that focuses on distributing a reinforcement learning algorithm to train on a large CPU cluster
Continue reading “Solving Atari games with Distributed Reinforcement Learning”

Using AI to help aging populations live better — microbiome may be the key

Using #AI to help Aging Populations Live Better  @UCSDnews  #IBMResearch

  • To help us do so, UC San Diego is launching a collaboration with IBM Research to enhance quality of life and independence for aging populations.
  • The new Artificial Intelligence for Healthy Living Center (AIHL), located on our La Jolla campus, will combine the technology, cognitive and life sciences strengths of IBM and UC San Diego.
  • At UC San Diego, we’re bringing together top researchers from across our campus including computer science and engineering, the San Diego Super Computer Center, cognitive science, the Center for Microbiome Innovation, medicine including psychiatry, and the Qualcomm Institute for Telecommunications and Information Technology.
  • Hands-on research experience with UC San Diego and IBM scientists will be complemented by jointly taught classes and seminars, and will showcase opportunities for trainee and faculty projects.
  • This project is supporting 15 UC San Diego faculty members, along with 46 students, postdocs and research staff.

UC San Diego and IBM Research will use AI to study the impact of genetics and the human microbiome on the cognition of older adults.
Continue reading “Using AI to help aging populations live better — microbiome may be the key”

Insight Artificial Intelligence Fellows Program

Announcing the Insight Artificial Intelligence Fellows Program in Silicon Valley & NYC

  • Insight alumni are shaping the future of the data industry
  • Join an elite professional network of data scientists, data engineers, and AI professionals
  • Solve real world problems with AI
  • Strong motivation to apply state-of-the-art AI methods to industry’s most difficult problems
  • Insight works with every Fellow until they land an artificial intelligence role

An intensive 7 week professional training fellowship leading to a career in artificial intelligence. Join top applied AI teams working with deep learning.
Continue reading “Insight Artificial Intelligence Fellows Program”