Game-independent AI agents for playing Atari 2600 console games

By Yavar Naddaf. 2010

46 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Reinforcement Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Sergey Levine
2. Jeff Clune
3. Pieter Abbeel
4. Kenneth O. Stanley
5. Shimon Whiteson
6. Joel Lehman
7. Marc G. Bellemare
8. J. Urgen Schmidhuber

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Trevor Darrell
3. Andrea Vedaldi
4. Joel Lehman
5. Jitendra Malik
6. Peter Stone
7. Andrew Zisserman
8. Dacheng Tao

In Reinforcement Learning

1. Marc G. Bellemare
2. Sergey Levine
3. Matthew J. Hausknecht
4. Michael H. Bowling
5. Jeff Clune
6. Pieter Abbeel
7. Shimon Whiteson
8. Peter Stone

Citations in the corpus

• Investigating Contingency Awareness Using Atari 2600 Games (2012, methodology): “The Basic method, derived from BASS (Naddaf 2010), encodes the presence of colors on the Atari screen. The background is first removed using a histogram method, described in Naddaf’s work (2010). Each game background is precomputed offline, using a sample trajectory containing 18,000 observations co”
• The Arcade Learning Environment: An Evaluation Platform for General Agents (2012, methodology): “Of these sets of features, BASS, DISCO and RAM were originally introduced by Naddaf (2010), while the rest are novel.”
• Sketch-Based Linear Value Function Approximation (2012, methodology): “The Atari domain was proposed as a game-independent platform for AI research by Naddaf [17].”
• Deep Apprenticeship Learning for Playing Video Games (2015, methodology): “To collect training data we modified the Arcade Learning Environment to record states (video frames) and actions (button presses) while a human plays the game.”
• A Comparison of learning algorithms on the Arcade Learning Environment (2014, background): “Naddaf (2010) presents some tables comparing decay and learning rates for 4 of the training set games.”
• Regularization in reinforcement learning (2011, unknown): “, 2007], and Atari 2600 console games [Naddaf, 2010].”
• Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents (2017, methodology): “For the behavior characterization, we follow an idea from Naddaf (2010) and concatenate Atari game RAM states for each timestep in an episode.”
• A Neuroevolution Approach to General Atari Game Playing (2014, methodology): “The Atari domain was first used as a research platform [11], then modified to interface with learning agents [3], [33], and released as ALE.”
• HyperNEAT-GGP: a hyperNEAT-based atari general game player (2012, methodology): “Results are averaged across five runs of HyperNEAT-GGP evolution and compared with previous results obtained using gradient descent Sarsa(λ) with linear function approximation[11].”