mab
mab copied to clipboard
Library for multi-armed bandit selection strategies, including efficient deterministic implementations of Thompson sampling and epsilon-greedy.
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use
Add custom errors so callers can handle different error modes: - reward source unreachable - probability calculation failed - arm selection failed
# Problem There is a newer version of Go available. # Solution Update `go.mod` to have the oldest supported version in use