UNIT 016 h allocated
Deep value learning
- DQN
- replay buffer
- target network
- exploration schedule
← Course catalogue
RL & ROBOTICS / SPECIALIST
RL310Deep reinforcement learning: DQN & policy gradients
24 h productive allocation9 h 36 m learning + 14 h 24 m training4 units · 1 assessed practical brief
Detailed course plan and training brief. A complete taught course pack is in preparation. The linked foundation module, where shown, is a smaller available resource.
BEFORE YOU BEGIN
A clear starting point.
Specialist studio; environment and faculty must be confirmed before scheduling
WHAT YOU SHOULD BE ABLE TO DEMONSTRATE
- Explain the assumptions and failure modes in deep value learning.
- Complete the practical: reproduce a small simulated DQN benchmark and report outcomes across seeds against a random policy.
- Defend the resulting training and evaluation config with reproducible evidence.
THE SUBJECTS / FOUR CONNECTED UNITS
What you will study.
UNIT 026 h allocated
Policy learning
- Policy gradients
- advantage
- actor-critic
- PPO concepts
UNIT 036 h allocated
Stable experimentation
- Seed variation
- termination handling
- evaluation episodes
- compute budgets
UNIT 046 h allocated
Transfer limits
- Simulator mismatch
- reward hacking
- safety constraints
- offline-policy limits
THE TRAINING STUDIO
Make the subject
visible in your work.
Reproduce a small simulated DQN benchmark and report outcomes across seeds against a random policy.
Evidence to produce
- Training and evaluation config
- return distributions
- failure episodes
Training brief for supervised delivery. Code, datasets, answer keys and environment acceptance must be completed in the full course pack before delivery is confirmed.
ASSESS THE REASONING
A question you
should be able to answer.
Why can one lucky learning curve hide an unstable algorithm?
Review dimensions
Technical correctness · reproducibility · interpretation of results · failure analysis · individual explanation. Course-level rubric and faculty moderation form part of the final adopted teaching pack.
External reference access, hardware, cloud accounts and paid services are separate. No third-party certification is included by implication.