Engage in a thought-provoking debate between Tomer Ullman and Laura Schulz on the development of intelligence. Explore key concepts such as the importance of theories, inferring good theories, and the grammar for theories. Delve into the process of finding ideal theories and proposing alternatives, with a focus on stochastic search and its application to children's learning. Examine the challenges of stochastic search, including the infinite search space and its limitations in utilizing existing knowledge and abilities. Investigate the nature of generating new ideas and the role of goal-oriented hypothesis generation and imagination in the development of intelligence.
Development of Intelligence - Tomer Ullman vs. Laura Schulz Debate
Overview
Syllabus
Intro
Prologue - What is this about?
Outline of Discussion
What Good is a Theory?
Inferring Good Theory
Grammar for Theories
Finding a Good Theory (ideal level)
Proposing Alternative Theories
Stochastic Search and Children
Mid-Summary 11
Handoff to Laura
What's wrong with stochastic search?
Problem 1: Even with prior knowledge, templates, and a hias towards simplicity, the search space is infinite.
Problem 1: Even with prior knowledge, templates, and a bias towards simplicity, the search space is infinite
Problem 2: Stochastic search does not make use of knowledge and abilities we seem to have
What does it mean to think of a new idea?
Goal-oriented hypothesis generation and imagination
Taught by
MITCBMM
