> ML_ALGORITHM // Q-LEARNING-TABULAR_v1.0
Tabular Q-Learning
Foundational model-free off-policy reinforcement learning algorithm that iteratively learns optimal action-value functions via temporal difference updates.
Value-Based Model-Free RLreinforcement-learninghigh-intrinsicmedium (1k-100k)
Back to All AlgorithmsComputational Complexity
Training Complexity:O(episodes * steps)
Inference Complexity:O(|A|) argmax table lookup
Hardware Profile
CPU Friendly:Yes
Requires GPU:No
Memory Footprint:low
Interpretability & Data
Interpretability Tier:high-intrinsic
Training Data Needs:medium (1k-100k)
Interpretability Assessment
Q-table values directly express expected cumulative discounted future returns per action.
Suitable Tasks & Supported Modalities
Suitable Tasks:
reinforcement learningdiscrete control
Supported Modalities:
tabular
Implementing Libraries
gymnasium
torchrl
Foundational Literature
Q-learningChristopher J. C. H. Watkins, Peter Dayan (1992) · Machine Learning
Common Pitfalls & Warnings
- Inability to generalize across unseen states; table size scales exponentially with state variables
- Overestimation of action values due to the max operator