Learning in Worlds with Objects

Slides:



Advertisements
Similar presentations
1 Knowledge Representation Introduction KR and Logic.
Advertisements

Cognitive Systems, ICANN panel, Q1 What is machine intelligence, as beyond pattern matching, classification and prediction. What is machine intelligence,
Presentation on Artificial Intelligence
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
Perception and Perspective in Robotics Paul Fitzpatrick MIT Computer Science and Artificial Intelligence Laboratory Humanoid Robotics Group Goal To build.
A Summary of the Article “Intelligence Without Representation” by Rodney A. Brooks (1987) Presented by Dain Finn.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
Intelligent Agents revisited.
Programming Fundamentals (750113) Ch1. Problem Solving
Artificial Intelligence
8/9/20151 DARPA-MARS Kickoff Adaptive Intelligent Mobile Robots Leslie Pack Kaelbling Artificial Intelligence Laboratory MIT.
Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Cognitive Robots © 2014, SNU CSE Biointelligence Lab.,
Wilma Bainbridge Tencia Lee Kendra Leigh
Operations and Equations Unit of Study: More Addition & Subtraction Strategies Global Concept Guide: 2 of 3.
Wheeler Lower School Mathematics Program Grades 4-5 Goals: 1.For all students to become mathematically proficient 2.To prepare students for success in.
CS 462: Introduction to Artificial Intelligence This course advocates the physical-symbol system hypothesis formulated by Newell and Simon in It.
Knowledge representation
New SVS Implementation Joseph Xu Soar Workshop 31 June 2011.
Towards Cognitive Robotics Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Christian.
Declarative vs Procedural Programming  Procedural programming requires that – the programmer tell the computer what to do. That is, how to get the output.
Intellectual Development of Toddlers (1-3)
The Sciences of the Artificial Herbert A. Simon Prefaces & Chapter 1.
University of Windsor School of Computer Science Topics in Artificial Intelligence Fall 2008 Sept 11, 2008.
1 The main topics in AI Artificial intelligence can be considered under a number of headings: –Search (includes Game Playing). –Representing Knowledge.
Introduction to Artificial Intelligence CS 438 Spring 2008 Today –AIMA, Ch. 25 –Robotics Thursday –Robotics continued Home Work due next Tuesday –Ch. 13:
Learning in the Large Information Processing Technology Office Learning Workshop April 12, 2004 Seedling Overview Learning in the Large MIT CSAIL PIs:
Artificial Intelligence: Research and Collaborative Possibilities a presentation by: Dr. Ernest L. McDuffie, Assistant Professor Department of Computer.
We believe that children's engineering can and should be integrated into the material that is already being taught in the elementary classroom -it does.
Some Thoughts to Consider 5 Take a look at some of the sophisticated toys being offered in stores, in catalogs, or in Sunday newspaper ads. Which ones.
#1 Make sense of problems and persevere in solving them How would you describe the problem in your own words? How would you describe what you are trying.
Artificial Intelligence Knowledge Representation.
Learning Procedural Knowledge through Observation -Michael van Lent, John E. Laird – 인터넷 기술 전공 022ITI02 성유진.
Lecture 14. Recap Problem Solving GA Simple GA Examples of Mutation and Crossover Application Areas.
Functionality of objects through observation and Interaction Ruzena Bajcsy based on Luca Bogoni’s Ph.D thesis April 2016.
NTT-MIT Collaboration Meeting, 2001Leslie Pack Kaelbling 1 Learning in Worlds with Objects Leslie Pack Kaelbling MIT Artificial Intelligence Laboratory.
Material design  Contextual factors  Purpose: teaching or testing?  Recycling  Exercises or tasks?  Communicative Language Teaching  Input: Data.
Introduction to Machine Learning, its potential usage in network area,
Are you a Left Brain or Right Brain?
Knowledge Representation Techniques
PRINCIPLE I. PROVIDE MULTIPLE MEANS OF REPRESENTATION
OPERATING SYSTEMS CS 3502 Fall 2017
What is cognitive psychology?
Algorithms and Problem Solving
Algorithms, Part 1 of 3 The First step in the programming process
Learning Fast and Slow John E. Laird
Artificial Intelligence and Lisp TDDC65
Chapter 5: Representing Knowledge
Lecture #1 Introduction
CS 4700: Foundations of Artificial Intelligence
Learning and Perception
Done Done Course Overview What is AI? What are the Major Challenges?
Artificial Intelligence
Algorithms I: An Introduction to Algorithms
Intelligent Adaptive Mobile Robots
CH. 1: Introduction 1.1 What is Machine Learning Example:
Development and Theorists
© James D. Skrentny from notes by C. Dyer, et. al.
Piedmont K-5 Math Adoption
Learning through Designing
Course Instructor: knza ch
We believe that children's engineering can and should be integrated into the material that is already being taught in the elementary classroom -it does.
Meeting Students Where They Are…
Programming Fundamentals (750113) Ch1. Problem Solving
Dr. Unnikrishnan P.C. Professor, EEE
#1 Make sense of problems and persevere in solving them
CS 416 Artificial Intelligence
CHAPTER I. of EVOLUTIONARY ROBOTICS Stefano Nolfi and Dario Floreano
Using Decision Structures
Guided Math.
Habib Ullah qamar Mscs(se)
Presentation transcript:

Learning in Worlds with Objects Leslie Pack Kaelbling MIT Artificial Intelligence Laboratory With Tim Oates, Natalia Hernandez, Sarah Finney

What is an Agent? Environment Observation Action A system that has an ongoing interaction with an external environment household robot factory controller web agent Mars explorer pizza delivery robot Environment Observation Action I’ll use my cartoon robot, Robby, as an example throughout the talk The name Robby is a reference to some old movie Sometimes the observation is the same as the state (not common, but a good starting point)

Agents Must Learn Learning is a crucial aspect of intelligent behavior human programmers lack required knowledge agents should work in a variety of environments agents should work in changing environments What to learn? World dynamics: What happens when I take a particular action? Reward: What world states are good? Housecleaning robot should be able to adapt to their adopted homes. Also, cope with rearrangements of furniture, etc...

Crisis Current state-of-the-art learning methods will not work in domains with multiple objects: These are crucial domains for robots of the future. ?

Representation Learning requires some sort of representation of states of the world. The choice of representation affects what information can be represented what kinds of generalizations the agent can make

Attribute Vector State-of-the-art representation for learning temperature = 48.2 pressure = 57.9 mB valve1 = open valve2 = closed time = 10:48AM backlog = 78 volume = 32.2 production = 45.5 …

Generalization over Attribute Vectors x time temp > 22 temp pressure < 3 time < 10AM close valve open valve add reagent increase temp

Complex Everyday Domains Attribute vector is impossibly big book1-on-book2: true book2-on-book1: false pen-is-yellow: true pen-is-blue: false lamp-on: true lamp-off: false ink-bottle-level: 50% lamp-in-bottle: false bottle-on-lamp: false paper1-color: gray paper2-color: white fabric-behind-lamp: true book2-is-clear: false book4-is-clear: false book1-is-clear: true block1-on-block2: false block3-unstable: true block2-on-table: false block1-in-front-of-lamp: true …

Generalization over Objects If book1 is on book2 and I move book2, then book1 will move If the cup is on the table and I move the table, then the cup will move If the pen is on the paper and I move the paper, then the pen will move If the coat is on the chair and I move the chair, then the coat will move For all objects A and B: If A is on B and I move B, then A will move

Referring to Objects Traditional symbolic AI has the problem of “symbol grounding”: How do I know what object is named by book1? on(book1,book2)

Deictic Expressions “Deixis” is Greek for “pointing” ima koko koko : here ima : now watashi-ga motteiru hako: the box I am holding watashi-ga motteiru hako: the box I am looking at watashi-ga motteiru hako watashi-ga miteiru hako

Automatic Generalization If I have an object in my hand and I open my hand, then the object that was in my hand is now on the table This is true, no matter what object is in your hand.

Communicating with Humans Natural language communication speaks of the world in terms of objects and their relationships uses deictic expressions Our robots of the future will have to be able to understand and generate human descriptions of the world

Long-Term Research Goal A robotic system with hand and cameras that can learn to achieve tasks efficiently through trial and error acquire natural language descriptions of the objects and their properties through “conversation” with humans

Short-Term Research Plan Explore deictic, object-based representation for learning algorithms build simulated hand-eye robot system that manipulates blocks (with real physics) have simulated robot learn to carry out tasks from trial and error Demonstrate empirically and theoretically that deictic representation is crucial for efficient learning

First Example Domain Unreliable block stacking: robot is rewarded for making tall piles of blocks the taller a pile is, the more likely it is to fall over when another block is added a pile can be made more stable by building piles to its sides Once the robot learns to do this task, keep the physics of the domain the same, but reward a more complex behavior.

Learning by Doing Having an initial task to perform focuses the robot’s attention on aspects of the environment Use extension of Utree learning algorithm to select important aspects of the environment Generate new deictic expressions dynamically: the-block-on-top-of(the-block-I-am-looking-at) Extend reinforcement learning methods to apply to object-based representations

Extracting General Rules There are too many facts that are true in any interesting environment. Solving tasks focuses attention on particular objects (named with deictic expressions) particular properties of those objects These objects and properties are likely of general importance: use them as input to association-rule learning algorithm to learn facts like: The thing that is on the thing that I am holding will probably fall off if I move

Enabling Planning Given general rules, the agent can “think” about the consequences of its actions and decide what to do, rather than learn through trial and error.

In Future An ambitious research project vision algorithms for learning segmentation and object recognition learning good properties and relations for characterizing the domain (“concept learning”) connect with natural language learning for word meanings

Don’t miss any dirt!