SORT 2025

Introduction

Introduction

2 player game: nine men’s morris 1 player: sliding game. solitaire? rubik’s cube

Reinforcement learning: aixi. Application of solomonoff?

puzzles: + rubic + wordle + tower of hanoi

goal agents v util agents. just find solution v find best solution

reinforcement learning split out: + q learning + temporal difference learning

Big h3: Decision theory under certainty

Certainty

for decision theory tree search: can need to find solution (ie pay off is 1 or 0) or "best" solution. if there are time/memory constraints, may not want to wait for best. may not know what best is until fully exhausted

Big h3: Intertemportal decision theory under certainty

Big h3: Decision theory under known uncertainty

Big h3: Decision theory under unknown uncertainty

Big h3: Game theory under certainty

sequential

divide dollar Ultimatum game

simulataenous games

keysian beauty contest + Prisoner’s dilemam

deep sequential

forward pruning (only look at best moves)

tables for start and end game eg in chess

Big h3: Game theory under uncertainty

in game theory, note that it is not stationary, so different to MDP and POMDP

Big h3: Robotics

SAT planning

planning graph

section on scheduling

critial path, critical path method

nlp

textual entailment in NLP.

Does one statement imply another, negate another, irrelavent to another?

task: assessing textual entailment

assessing semantic similarity

sentiment analysis

machine translation