Overview
This course will introduce the probabilistic approach to cognitive science, in which learning and reasoning are understood as inference in complex probabilistic models. Examples will be drawn from areas including concept learning, causal reasoning, social cognition, and language understanding. Formal modeling ideas and techniques will be discussed in concert with relevant empirical phenomena.
Instructor:
Noah Goodman (ngoodman at stanford dot edu)
TA:
Long Ouyang (louyang at stanford dot edu).
Meeting time: T Th 2:30-4.
Meeting place: building 200, room 202 (in the history
corner of the quad).
Office hours: Long Ouyang will hold office hours Friday 1-2 in building 420 (jordan hall), room 330.
Discussion: In addition we are trying out an online discussion tool:
Piazza. Post your questions there, and answer others' questions (it will be considered in your participation grade!).
Assignments and grading
Students (both registered and auditing) will be expected to do assigned readings before class.
Registered students will be graded based on:
- 30% Class participation (including exercises, paper presentation, and Piazza activity).
- 35% Homework.
- 35% Final project (project instructions).
Readings
Readings for each week will be linked from the calendar section. (In some cases these will require an SUNet ID to access. See the instructor in case of trouble.) There will be three main sources of readings:
- The probabilistic models of cognition wiki.
- Draft chapters from the probabilistic models of cognition book (by Griffiths, Tenenbaum, Chater, Kemp, Goodman, Yuille). These chapters are often very rough drafts; feedback is welcome.
- Selected research papers.
Pre-requisites
There are no formal pre-requisites for this class. However, the course will move relatively quickly and have technical content. Students should be already familiar with the basics of probability and programming (or be willing to learn this background as needed).