HyperNEAT-GGP: a hyperNEAT-based atari general game player

By Matthew J. Hausknecht, Piyush Khandelwal, Risto Miikkulainen, Peter Stone. 2012

65 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research, Reinforcement Learning, Neural Architecture Search & AutoML

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. Pieter Abbeel
3. J. Urgen Schmidhuber
4. Marc G. Bellemare
5. Peter Stone
6. Matthew J. Hausknecht
7. Yoshua Bengio
8. Shimon Whiteson

In Management Science and Operations Research

1. Marc G. Bellemare
2. Michael H. Bowling
3. Peter Stone
4. Joel Veness
5. Satinder Singh
6. Tom Schaul
7. Matthew J. Hausknecht
8. Risto Miikkulainen

In Reinforcement Learning

1. Sergey Levine
2. Pieter Abbeel
3. Marc G. Bellemare
4. Matthew J. Hausknecht
5. Michael H. Bowling
6. Satinder Singh
7. Peter Stone
8. Honglak Lee

In Neural Architecture Search & AutoML

1. Joel Lehman
2. Peter Stone
3. Marc G. Bellemare
4. Joel Veness
5. Michael H. Bowling
6. David R Ha
7. Matthew J. Hausknecht
8. Risto Miikkulainen

Citations in the corpus

• The Arcade Learning Environment: An Evaluation Platform for General Agents (2012, methodology): “he games of Pong and Frogger, using their own simulator. Their proposed method takes advantage of human trajectories to identify state features that are important for playing console games. Recently, Hausknecht et al. (2012) proposed HyperNEAT-GGP, an evolutionary approach for nding policies to play”
• A Neuroevolution Approach to General Atari Game Playing (2014, background): “playing agent [18] demonstrated to learn on two different Atari games: Freeway and Asterix.”
• Image Classification Using Generative Neuro Evolution for Deep Learning (2015, background): “HyperNEAT has shown promise in simple visual discrimination tasks [5, 3, 26, 10], but has been limited to two or fewer hidden layers.”
• Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning (2014, background): “Specifically, low-level features are learned to encode low-level statistical dependencies (e.g., “edges” in images), and higher-level features encode higher-order dependencies of the lower-level features (e.g., “object parts”) [14].”
• Sketch-Based Linear Value Function Approximation (2012, methodology): “Different learning methods have been evaluated on the Atari 2600 platform [9, 26, 12].”
• General Video Game Playing (2013, background): “has investigated the use of evolutionary neural networks for creating general video games players (6).”
• Neuroevolution of self-interpretable agents (2020, methodology): “popularity of deep RLbased approaches for vision-based tasks, indirect encoding methods from the neuroevolution literature have been used to tackle challenging vision-based tasks such as Atari domain [43] and car navigation from pixels [56]. Indirect encoding methods are inspired by biological genot”
• Bayesian Learning of Recursively Factored Environments (2013, background): “agents that attempt to perform well over a large set of domains (Bellemare et al., 2013; Hausknecht et al., 2012).”
• Generative NeuroEvolution for Deep Learning (2013, background): “HyperNEAT has shown promise in simple visual discrimination tasks [13, 14, 15, 16], but has been limited to two or fewer hidden layers.”