You will implement the core Q-learning logic while the scaffolding (seeding, BFS reachability, environment helpers, printing utilities) is provided for you. Parts Part 1 (Training): Learn for

COE292 — Homework 1: Intelligent Explorer (Starter)

Complete the TODO sections only. Do not rename functions or change their arguments.

You will implement the core Q-learning logic while the scaffolding (seeding, BFS reachability, environment helpers, printing utilities) is provided for you.

Parts

• Part 1 (Training): Learn for 1000 episodes from zero knowledge; print knowledge matrices and the final greedy path + energy.
• Part 2 (Step-by-Step): Apply Q-updates along a fixed path with a given initial knowledge vector; print per-step updates and the final matrix.

Rules

• Only the Python standard library and numpy are allowed.
• Keep output formats exact (spacing/labels) because the autograder parses them.
• Set your STUDENT_ID in the first code cell.

Expert Help With This Assignment — On Your Terms