Module: tf_agents.bandits.agents.exp3_agent

Implements the EXP3 bandit algorithm.

Implementation based on

"Bandit Algorithms" Lattimore and Szepesvari, 2019 https://tor-lattimore.com/downloads/book/book.pdf

Classes

class Exp3Agent: An agent implementing the EXP3 bandit algorithm.

Functions

exp3_update_value(...)

selective_sum(...): Sums entries in values, partitioned using partitions.

absolute_import Instance of __future__._Feature
division Instance of __future__._Feature
print_function Instance of __future__._Feature