Sliding Window Ladder¶
Who This Is For¶
Use this ladder when a problem sits between DP and window techniques, and the main challenge is maintaining the right bounded state while scanning.
Warm-Up¶
- variable-window reasoning
- weighted windows over sorted structure
Core¶
- combine window invariants with DP-style state interpretation
- avoid recomputing window contributions from scratch
Retrieval Layer¶
- moving minimum / best recent transition -> monotonic-deque-min.cpp
- balanced lower median in a dynamic window -> sliding-median-two-multisets.cpp
- short reminder sheet -> DP cheatsheet
- exact monotone-window rep -> Sliding Window Minimum
Example Notes¶
Exit Criteria¶
You are ready to move on when you can:
- explain what the active window means at every step
- tell whether a problem is really two-pointers, really DP, or a hybrid
- maintain weighted contributions incrementally
- move between geometric preprocessing and windowed counting without losing the model