Skip to content

Berkeley CS285 Homework and Final Projects: The CPU, GPU, and H100 Boundary

Aug 22, 2026 1 min
TL;DR 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.
Table of Contents
  1. Compute ledger
  2. A three-tier route
  3. Choosing a final project
  4. Delivery boundary
  5. References

🌏 中文版

The public Spring 2026 starter repository contains HW1–5 and code for two default final projects under an MIT license. That makes independent implementation possible.

The series overview owns the complete public-versus-enrolled access boundary. This article owns compute and project tradeoffs.

Compute ledger

WorkImplementation focusOfficial compute signalSelf-study choice
HW1MSE, DAgger, flow-matching imitationREADME found local CPU faster in testingComplete on CPU
HW2Policy gradients, reward-to-go, baselineSeconds to about ten minutes per runComplete on CPU; add seeds
HW3DQN and SACMsPacman and HalfCheetah may take about three GPU hoursDo small environments; choose one expensive task
HW4REINFORCE, GRPO, LLM rewardsModal wrapper defaults to H100; four required runsReduce format-copy first; set a cloud budget
HW5SAC+BC, IQL, FQL, OGBenchOne run may take about six hours; tuning accumulatesStart with one task and seed

These times are estimates for the official environments, not guarantees across hardware. Cloud pricing changes, so check current rates and define a stopping condition before launching paid resources.

A three-tier route

First, complete HW1 and HW2 and establish logging, evaluation, and multi-seed habits. Second, run HW3 on cheap environments and choose at most one expensive environment. Third, select one reduced experiment from each of HW4 and HW5, with a time or spending cap.

Record environment versions, seeds, commands, wall-clock time, hardware, and failed runs. Saving only the best curve discards the reproducibility lesson that matters most in RL.

Choosing a final project

The public final project outline accompanies default Offline-to-Online RL and LLM RL options. The first extends a fixed dataset with limited interaction. The second involves preference data, reward modeling, and policy optimization. Course documents mention enrolled-student Modal resources and an H200 option; neither is a public entitlement.

A self-study project can be one baseline, one modification, three seeds, one principal plot, and a limitations section. Without a GPU, choose a small simulated offline-to-online setting rather than assuming LLM training costs.

Delivery boundary

The public path can reproduce a self-evaluated version of the assignments. It cannot claim completion of the Berkeley credit-bearing course. The series overview maintains the full list of video, discussion, grading, and course-support limits.

References