Alternative names (2)
ruleChange • ruleModificationRandom triples
| Subject | Object |
|---|---|
| gptkb:Q-learning_algorithm | Q(s,a) ← Q(s,a) + α [r + γ max Q(s',a') - Q(s,a)] |
| gptkb:TD(0) | V(s) ← V(s) + α [r + γ V(s') − V(s)] |
| gptkb:Self-Organizing_Map | neighborhood function |
| gptkb:Least_Mean_Squares_(LMS) | w(n+1) = w(n) + μ e(n) x(n) |
| gptkb:LMS_algorithm | w(n+1) = w(n) + μ e(n) x(n) |
| gptkb:Flag_Football_Under_14 | flag pulling instead of tackling |
| gptkb:Stochastic_gradient_descent | parameter update using single or few samples |
| gptkb:Global_Rapid_Rugby | faster restarts |
| gptkb:Adam_Optimizer | element-wise parameter updates |
| gptkb:Least_Mean_Squares | w(n+1) = w(n) + μ e(n) x(n) |
| gptkb:CPISRA_Football_7-a-side_World_Championships | smaller field |
| gptkb:SARSA | Q(s,a) ← Q(s,a) + α [r + γ Q(s',a') − Q(s,a)] |
| gptkb:Adam_Optimizer | bias-corrected estimates |
| gptkb:WCST | rules change without warning |
| gptkb:2015_NLDS_slide | led to new slide rule (Utley Rule) |
| gptkb:self-organizing_maps | weight adaptation based on neighborhood |
| gptkb:L-BFGS_algorithm | approximates inverse Hessian matrix |
| gptkb:CPISRA_Football_7-a-side_World_Championships | throw-ins allowed with one hand |
| gptkb:RMSprop | divides learning rate by moving average of squared gradients |
| gptkb:Elo_rating_system | rating changes after each game |