Data Skeptic [MINI] Markov Chains
5 snips
Mar 20, 2015 This podcast discusses Markov Chains and their applications in various systems including stop lights, text prediction, and bowling. The hosts explore the concept of Markov Chains in daily life and technology, as well as their impact on partially observable state spaces.
AI Snips
Chapters
Transcript
Episode notes
Tic-Tac-Toe Illustrates Memoryless State
- The hosts use Tic-Tac-Toe to show that order of past moves doesn't always matter for the current state's next player.
- You can deduce whose turn it is from counts of Xs and Os without knowing move order.
Markov Assumption Simplifies State Dependence
- The Markov assumption says the current state depends only on the previous state and intervening events.
- This simplifies modeling stochastic systems by ignoring longer histories when the present state fully summarizes past information.
Monopoly Shows Fully Observable States
- Monopoly is used to show a fully observable state where new players can pick up a game from the board.
- Seeing properties, money, and piece positions gives enough information to continue play without prior history.