GPTKB
Browse
Query
Compare
Download
Publications
Contributors
Search
updateRule
URI:
https://gptkb.org/prop/updateRule
79
triples
GPTKB property
Alternative names (2)
ruleChange
•
ruleModification
Random triples
Subject
Object
gptkb:Adam_Optimizer
element-wise parameter updates
gptkb:Kohonen_map
winner-takes-all
gptkb:RMSprop
divides learning rate by moving average of squared gradients
gptkb:Particle_Swarm_Optimization
position update
gptkb:L-BFGS_algorithm
approximates inverse Hessian matrix
gptkb:Q-Learning
gptkb:Bellman_equation
gptkb:REINFORCE_Algorithm
gradient ascent
gptkb:Adam_Optimizer
uses first and second moment estimates
gptkb:AdaGrad
per-parameter learning rate
gptkb:WCST
rules change without warning
gptkb:Q-learning
gptkb:Bellman_equation
gptkb:Global_Rapid_Rugby
no kicking out on the full
gptkb:Para_7-a-side_National_Team
no offside rule
gptkb:Stochastic_gradient_descent
parameter update using single or few samples
gptkb:REINFORCE_algorithm
gradient ascent
gptkb:Para_7-a-side_National_Team
throw-ins can be one-handed
gptkb:Exponentially_Weighted_Recursive_Least_Squares
recursive
gptkb:Conway's_Game_of_Life
synchronous
gptkb:Global_Rapid_Rugby
faster restarts
gptkb:Self-Organizing_Map
weight adaptation