The acquisition of inductive constraints

نویسندگان

  • Charles Kemp
  • Joshua Tenenbaum
  • Matthew Wilson
چکیده

Human learners routinely make inductive inferences, or inferences that go beyond the data they have observed. Inferences like these must be supported by constraints, some of which are innate, although others are almost certainly learned. This thesis presents a hierarchical Bayesian framework that helps to explain the nature, use and acquisition of inductive constraints. Hierarchical Bayesian models include multiple levels of abstraction, and the representations at the upper levels place constraints on the representations at the lower levels. The probabilistic nature of these models allows them to make statistical inferences at multiple levels of abstraction. In particular, they show how knowledge can be acquired at levels quite remote from the data of experience—levels where the representations learned are naturally described as inductive constraints. Hierarchical Bayesian models can address inductive problems from many domains but this thesis focuses on models that address three aspects of high-level cognition. The first model is sensitive to patterns of feature variability, and acquires constraints similar to the shape bias in word learning. The second model acquires causal schemata—systems of abstract causal knowledge that allow learners to discover causal relationships given very sparse data. The final model discovers the structural form of a domain—for instance, it discovers whether the relationships between a set of entities are best described by a tree, a chain, a ring, or some other kind of representation. The hierarchical Bayesian approach captures several principles that go beyond traditional formulations of learning theory. It supports learning at multiple levels of abstraction, it handles structured representations, and it helps to explain how learning can succeed given sparse and noisy data. Principles like these are needed to explain how humans acquire rich systems of knowledge, and hierarchical Bayesian models point the way towards a modern learning theory that is better able to capture the sophistication of human learning. Thesis Supervisor: Joshua Tenenbaum Title: Associate Professor of Cognitive Science

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Evaluating the Effectiveness of Deductive and Inductive Form-Focused In-struction on Iranian EFL Learners' Implicit and Explicit Knowledge of Non-Generic Definite Article

This study investigated the relative effects of deductive and inductive form-focused instruction (FFI) on the acquisition of 4 non-generic definite article uses(cultural, situational, structural, and textual) as assessed by explicit and implicit outcome measures. The tests utilized to assess EFL learners' acquisition of definite article uses were timed and untimed grammaticality judgment tests....

متن کامل

The Impact of Deductive, Inductive, and L1-Based Consciousness-Raising Tasks on EFL Learners' Acquisition of the Request Speech Act

The necessity and importance of teaching pragmatics has come to light by many researchers (e.g. Rose & Kasper, 2001). Due to the consensus over the need to teach pragmatic competence, the main issue now centers on the question of how we should teach this competence in the most effective way. Consistent with this line of research, the present study aimed to investigate the effectiveness of deduc...

متن کامل

Three-dimensional Magnetotelluric Modeling of data from Northeast of Gorgan Plain

Magnetotelluric measurements have been conducted in the period range of 0.005-128 s along five parallel east-west directed profiles including 85 sites totally in the north-eastern part of Gorgan Plain, Golestan Province, North of Iran; with the aim of exploring iodine. Distortion and dimensionality analysis of data imply the existence of a north-south elongated two-dimensional model with some l...

متن کامل

An Architecture for Knowledge Management in Intelligent Tutoring System

This paper discusses a general framework for knowledge acquisition and management in an intelligent tutoring system. This system is based on “Learning by performance errors” theory stating that in a given domain knowledge there is a set of constraints that must be satisfied in order to provide the correct solution to the problem. This paper addresses the issues of representing complex and gener...

متن کامل

Inductive Logic Programming : Theory And

> Inductive Logic Programming (ILP) is a new discipline which investigates the inductive construction of rst-order clausal theories from examples and background knowledge. We survey the most important theories and methods of this new eld. Firstly, various problem speciications of ILP are formalised in semantic settings for ILP, yielding a \model-theory" for ILP. Secondly, a generic ILP algorith...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007