Greedy vs dynamic programming: greedy makes locally optimal choices, DP considers all subproblems

Overlapping subproblems

Ever calculated a huge Fibonacci sequence by hand?

Viterbi semiring

How do we find the best path in a maze of choices?

P

P vs NP asks if every problem whose solution is quickly verifiable can also be quickly solved

Greedy vs beam search decoding: greedy picks best token, beam maintains k candidates

Ever wondered why Google Search sometimes shows you the top results first?

DPO simplifies: removes the explicit reward model, trains directly on preferences

DeFi removes intermediaries like banks

second-order methods (Newton's) converge faster but are expensive: O(n³) per step

Second-order methods converge faster due to quadratic convergence but are expensive due to O(n³) per iteration