Who is this for? This assessment checks whether you have the math, Python, NumPy, PyTorch, and basic RL concepts needed to start the curriculum comfortably.
New to programming? Start with the Learning path Phase 0. Unsure about math? Try the Math for RL track first. After that, use this assessment to see if you are ready for the curriculum.
25 questions to assess your foundational knowledge before the 100-chapter reinforcement learning curriculum. Answer honestly; each topic page has solutions and explanations to help you understand the “why,” not just the “what.” If you can answer at least 20 correctly and feel comfortable with the concepts, you are ready to start. If you struggled with many, review Prerequisites or the Learning path and come back.
Recommended order
Work through the topics in this order for a logical flow: math foundations → programming (Python, NumPy, PyTorch) → RL framework and value functions → tabular methods → function approximation and deep RL → self-assessment.
- Probability & statistics (Q1–Q2)
- Linear algebra (Q3–Q4)
- Calculus (Q5–Q6)
- Python basics (Q7)
- NumPy (Q8)
- PyTorch basics (Q9)
- RL framework (Q10–Q13)
- Value functions & Bellman (Q14–Q15)
- Tabular methods (Q16–Q19)
- Function approximation & Deep RL (Q20–Q24)
- Final self-assessment (Q25)
Syllabus at a glance
| Topic | What you’ll do | Questions |
|---|---|---|
| Probability & statistics | Sample mean, variance, expectation vs sample average, law of large numbers; bandit-style problems and code | Q1, Q2 |
| Linear algebra | Dot product, matrix-vector product, \(\nabla_w (Aw)\); NumPy snippet | Q3, Q4 |
| Calculus | Derivatives, chain rule, sigmoid; small code check | Q5, Q6 |
| Python basics | Moving average, list comprehensions, dict of returns | Q7 |
| NumPy | Create array, set row, element-wise product; slices and shapes | Q8 |
| PyTorch basics | Tensors, requires_grad, backward(), autograd examples | Q9 |
| RL framework | Agent, environment, state, action, reward; Markov; exploration-exploitation; \(\gamma\) | Q10–Q13 |
| Value functions & Bellman | \(V^\pi(s)\), \(Q^\pi(s,a)\); Bellman expectation equation; tiny MDP | Q14, Q15 |
| Tabular methods | Policy iteration, value iteration; MC vs TD; on-policy vs off-policy; Q-learning update | Q16–Q19 |
| Function approximation & Deep RL | Why FA; policy gradient update; ε-greedy, noisy nets; experience replay; actor-critic | Q20–Q24 |
| Final self-assessment | Rate comfort in Python, math, and RL; links to review | Q25 |
Each topic page includes worked problems with explanations, code examples with explanations, and math examples with step-by-step reasoning. Use them to fill gaps before starting the Curriculum.