Assumptions of Decision Making Models in AGI
Table of Contents
- AGI agents need to make decisions
- The Normative theory that studies decision making is Decision Theory.
- In Decision Theory we take action to maximize expected utility
- When we have to take sequence of decisions, it is called Markov Decision Process where we take actions to maximize exected total reward
- In AI, Decision theory and Reinforcement Learning is the best known [Page 1] approach to solve this problem
1. Assumption on Task - Actions are Infinite
It is unreasonable to assume that at any state, all possible actions are listed.
Actions in small scale may be discrete or a finite collection of distributions, but at the level where planning happens set of all possible actions is infinite.
Such actions can in principle to thought to be recursively composed of a set of basic operations/actions. But the decision making happens not at those basic actions but at level of composed actions.
- It is a task in itself to know what actions can be taken and what actions should we evaluate.
Thus decision making involves composing short timestep actions to get longer term action over which planning can be done.
i.e. decision making is often not about selection but selective composition. [Page 2]
Can we use intent (as in Intention Conditioned Value Function) to represent composed actions over which planning can be done?
How does Hierarchical Reinforcement Learning come into play?