- See more
See all on WikipediaTD-Gammon - Wikipedia
TD-Gammon is a computer backgammon program developed in 1992 by Gerald Tesauro at IBM's Thomas J. Watson Research Center. Its name comes from the fact that it is an artificial neural net trained by a form of temporal-difference learning, specifically TD-Lambda. The final version of TD-Gammon (2.1) was … See more
TD-Gammon's exclusive training through self-play (rather than tutelage) enabled it to explore strategies that humans previously had not considered or had ruled out erroneously. Its success with unorthodox strategies had a significant impact on the … See more
Unlike previous neural-net backgammon programs such as Neurogammon (also written by Tesauro), where an expert trained the program by supplying the "correct" evaluation … See more
Wikipedia text under CC-BY-SA license TD-Gammon algorithm - Medium
Jul 9, 2020 · TD-Gammon was designed as a way to explore the capability of multilayer neural networks trained by TD (λ) to learn complex nonlinear …
- Estimated Reading Time: 11 mins
This article presents a game-learning program called TD-Gammon. TD-Gammon is a neural network that trains itself to be an evaluation function for the game of backgammon by playing …
- File Size: 67KB
- Page Count: 16
dellalibera/td-gammon: TD-Gammon implementation - GitHub
Temporal difference learning and TD-Gammon | Communications …
Mar 1, 1995 · TD-Gammon is a neural network that trains itself to play backgammon by playing against itself and learning from the outcome. The program has surpassed all previous …
- Author: Gerald Tesauro
- Publish Year: 1995
TD-Gammon is a neural network that is able to teach itself to play backgammon solely by playing against itself and learning from the results, based on the TD(X) reinforcement learning …
- People also ask
TD-Gammon Explained - Papers With Code
TD-Gammon is a method that combines temporal difference learning and a feedforward neural network to play backgammon. Learn about its components, papers, results, and usage over time on Papers With Code.
Training in TD-Gammon •Initial feature representation was a raw encoding of board positions •NN was simple by today’s standards –40 hidden nodes •Main training paradigm was “self play” …
TD-Gammon, a Self-Teaching Backgammon Program, Achieves …
TD-Gammon is a neural network that is able to teach itself to play backgammon solely by playing against itself and learning from the results, based on the TD(λ) reinforcement learning …
fomorians/td-gammon: Implementation of TD …
Before DeepMind tackled playing Atari games or built AlphaGo there was TD-Gammon, the first algorithm to reach an expert level of play in backgammon. Gerald Tesauro published his paper in 1992 describing TD-Gammon as a …
- Some results have been removed
