From e5603da9a9d04b5db4f35e224b88a6044aaca780 Mon Sep 17 00:00:00 2001 From: Jens Ahrensfeld Date: Fri, 4 Apr 2025 16:36:07 +0200 Subject: [PATCH] refactored --- k-bandits.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/k-bandits.py b/k-bandits.py index 1cfd591..f5ccea5 100644 --- a/k-bandits.py +++ b/k-bandits.py @@ -78,11 +78,11 @@ def test(_k_arms: int, _episode_len: int, ql_star, _tie_break, _param: Param) -> _reward = _reward_z[j] + ql_star[_a] # calc - _nu[_a] = _nu[_a] + 1 + _nu[_a] += 1 if _param.alpha > 0: - _qu[_a] = _qu[_a] + _param.alpha * (_reward - _qu[_a]) + _qu[_a] += _param.alpha * (_reward - _qu[_a]) else: - _qu[_a] = _qu[_a] + (_reward - _qu[_a]) / _nu[_a] + _qu[_a] += (_reward - _qu[_a]) / _nu[_a] # Reduce tendency to explore with number of steps (or with age for humans) _epsilon = _param.epsilon * (1 - _param.rho)