Skip to content
All tags

#deep-reinforcement-learning

3 posts
aiguideReading Stanford CS234 Reinforcement Learning

DQN — the Deadly Triad, Experience Replay, and Fixed Q-Targets | Stanford CS234 Topic Guide

Q-learning converges with a table but can diverge once you add function approximation. CS234 blames the deadly triad: bootstrapping, function approximation, and off-policy learning all at once. DQN holds things together with two tricks. Experience replay breaks the correlation between consecutive samples, and fixed Q-targets keep the target still for C steps. In the Atari ablation table the slides show, Breakout goes from 3 with a linear model and 3 with a plain deep network to 317 with both tricks; replay alone reaches 241.

#cs234#ai-course#stanford#reinforcement-learning#deep-reinforcement-learning
learningguideReading Berkeley CS285 Deep Reinforcement Learning

The CPU, GPU, and H100 Boundary | Berkeley CS285 Topic Guide

Five assignments move from CPU-friendly imitation learning to H100-based LLM RL and six-hour offline-RL runs; self-learners should use three compute tiers instead of copying the entire enrolled workflow.

#cs285#berkeley#deep-reinforcement-learning#gpu#self-study
learningguideReading Berkeley CS285 Deep Reinforcement Learning

Deep Reinforcement Learning: From Decision Concepts to Policy Learning and Experiments | Berkeley CS285 Course Overview

Spring 2026 CS185/285 publishes slides for 25 lectures, nine discussion units, five assignments, and starter code; lecture recordings are public on the RAIL YouTube channel, but Ed, Gradescope, and office hours remain enrolled-only, while HW4 defaults to an H100, so this is not a zero-cost open course.

#cs285#berkeley#deep-reinforcement-learning#self-study#ai-course