Psychology 209 – Winter 2019 March 7, 2019

Slides:



Advertisements
Similar presentations
Pat Langley School of Computing and Informatics Arizona State University Tempe, Arizona USA Modeling Social Cognition in a Unified Cognitive Architecture.
Advertisements

Cognitive level of Analysis
Mathematics the Preschool Way
The Characteristics of an Experimental Hypothesis
CS 478 – Introduction1 Introduction to Machine Learning CS 478 Professor Tony Martinez.
Reasoning Abilities Slide #1 김 민 경 Reasoning Abilities David F. Lohman Psychological & Quantitative Foundations College of Education University.
Intelligence The Curious Case of Intelligence 13th Thinking Conference June 2007 Philip Adey King’s College London.
Knowledge and Memory: How we conceptualize information.
Section 1.1 What is Psychology?.
General Information Course Id: COSC6342 Machine Learning Time: TU/TH 10a-11:30a Instructor: Christoph F. Eick Classroom:AH123
1 PART I: PSYCHOANALYSIS & MOTIVES 2 INTRODUCTION TO THE STUDY OF MOTIVES DEFINITION & PROPERTIES CHARACTERISTICS OF MOTIVATED BEHAVIOR MOTIVES AND PERFORMANCE.
Overview of Cognitive Science for Teachers
Social Cognition Psych. 414 Prof. Jessica Sommerville.
CPS 270: Artificial Intelligence Machine learning Instructor: Vincent Conitzer.
2009 ML Project: Goal: Do some real machine learning… A project you are interested in works better Data is often the hard part (get it in plenty of time)
CSE5900 Lecture 9AI & MM Pt. 11 AI and MM Part 1 (An Embarrassingly Over-Simplified Introduction)
1 CSC111H User Interface Design Dennis Burford
Eliminativism Philosophy of Mind Lecture 5 (Knowledge and Reality)
Anomalous monism Michael Lacewing uk.
Is an Implicit Faith in IRT Warranted?. This “college level” math in 1769 is now the stuff of elementary school curricula everywhere in the world … can’t.
Copyright © 2014 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
Why Study Psychology?.  Nowhere, yet. Here we go!
Presented by:- Reema Tariq Artificial Intelligence.
Continuing the School Visit: Deepening the Next Level of Work East Boston High School December 9, 2014.
CSC-305 Design and Analysis of AlgorithmsBS(CS) -6 Fall-2014CSC-305 Design and Analysis of AlgorithmsBS(CS) -6 Fall-2014 Design and Analysis of Algorithms.
Foundation Degree in Business Victoria Hadfield
Evolvable dialogue systems
Software Defects Cmpe 550 Fall 2005
Creativity of Algorithms & Simple JavaScript Commands
Shared Intentionality
Deep Learning: What is it good for? R. Burgmann
What is cognitive psychology?
Supporting School Transformation
Evaluating your project
Interactive Topic Test
Dynamics of Learning & Distributed Adaptation
What is Psychology?.
Psychology 209 – Winter 2017 March 9, 2017
Goodfellow: Chap 1 Introduction
Draft for factual comments, suggestions on overlaps welcome.
Sociology: Understanding and Changing the Social World
Spring Courses CSCI 5922 – Probabilistic Models (Mozer) CSCI Mind Reading Machines (Sidney D’Mello) CSCI 7000 – Human Centered Machine Learning.
Epileptic Seizure Prediction
Videos NYT Video: DeepMind's alphaGo: Match 4 Summary: see 11 min.
Eliminative materialism
Lee, Jung-Woo Interdisciplinary Program in Cognitive Science
//Global Warming//.
Artificial Intelligence in Healthcare
Teaching with Instructional Software
Chapter 12: Automated data collection methods
Goodfellow: Chap 1 Introduction
Neuroscience-Inspired Artificial Intelligence
Psychology 209 – Winter 2018 March 8, 2018
Scientific Inquiry Standard B – 1.1.
Scientific Inquiry Standard B – 1.1.
Metaheuristic methods and their applications. Optimization Problems Strategies for Solving NP-hard Optimization Problems What is a Metaheuristic Method?
Intelligent Antenna Sharing in Wireless Networks
The Scientific Method REVIEW
Symbolic AI 2.0 Yi Zhou.
Cloud Computing Erasmus+ Project
Machines 1. What is Work?.
Improving Machine Learning using Background Knowledge
Manage testing by time boxes
Chapter 7: Learning.
Six Characteristics that apply to scientific concepts
Toward a Great Class Project: Discussion of Stoianov & Zorzi’s Numerosity Model Psych 209 – 2019 Feb 14, 2019.
Instructor: Vincent Conitzer
Thinking like a Scientist
This presentation was developed by Dr. Steven C
Morteza Kheirkhah University College London
Presentation transcript:

Psychology 209 – Winter 2019 March 7, 2019 Successes, Limitations, and Future Directions for Neural Network Models of Cognition Psychology 209 – Winter 2019 March 7, 2019

Topics and students who raised them

Lake et al Pattern recognition vs model building: Cognition is about using models to understand the world, to explain what we see, to imagine what could have happened that didn’t, or what could be true that isn’t, and then planning actions to make it so.

Start up software Intuitive physics Intuitive psychology Infants have primitive object concepts that allow them to track objects over time and allow them to discount physically implausible trajectories – e.g. they know that objects will persist over time and that they are solid and coherent. Intuitive psychology Infants understand that other people have mental states like goals and beliefs, and this understanding strongly constrains their learning and predictions.

Learning as model building Explaining observed data through the construction of causal models of the world. ‘Early present capacities for intuitive physics and psychology are also causal models of the world’. A primary job of learning is to extend and enrich these models and build analogous causally-structured theories of other domains. Human learning is richer and more efficient than state-of-the-art algorithms in machine learning Compositionality and learning to learn are ingredients that make this type of rapid model learning possible

Model Based and Model Free Methods Using a model is cumbersome and slow; model free reinforcement learning can allow real-time ‘control’. Humans combine MB and MF competitively and cooperatively

Two Challenges Characters Frostbite

One example: Omniglot Classification of new examples Generation of new examples Parsing an object into its parts Generation of new concepts from related examples

Lake et al’s solution

How might you address this challenge using neural networks?

DQN learns Frostbyte slowly – people can do well from brief instruction or from watching a good player Construct an igloo Jump on white ice flows Gather fish Don’t fall in the water Avoid geese & polar bears

How might you address this challenge using neural networks?

Emergent intelligence vs built-in intelligence May be easier to create Since it is designed, it is likely to be easier to understand You need to have just the right stuff to get the stuff you want to learn to fit within it May not deal with quasiregularity Emergent Not as easy to create Not as easy to understand Deals with quairegularity Involves less prior commitment to structure

What other challenges can you envision?