AAAI 2013 Spring Symposium

Lifelong Machine Learning

March 25-27, 2013
Stanford University, Stanford, CA

Part of the AAAI 2013 Spring Symposium Series

Overview

Topics

Invited Speakers

Schedule

Important Dates

Overview

Humans learn to solve increasingly complex tasks by continually building upon and refining knowledge over a lifetime of experience. This process of continual learning and transfer allows us to rapidly learn new tasks, often with very little training. Over time, it enables us to develop a wide variety of complex abilities across many domains.

Despite recent advances in transfer learning and representation discovery, lifelong machine learning remains a largely unsolved problem. Lifelong machine learning has the huge potential to enable versatile systems that are capable of learning a large variety of tasks and rapidly acquiring new abilities. These systems would benefit numerous applications, such as medical diagnosis, virtual personal assistants, autonomous robots, visual scene understanding, language translation, and many others.

Learning over a lifetime of experience involves a number of procedures that must be performed continually, including:

Discovering representations from raw sensory data that capture higher-level abstractions,
Transferring knowledge learned on previous tasks to improve learning on the current task,
Maintaining the repository of accumulated knowledge, and
Incorporating external guidance and feedback from humans or other agents.

Each of these procedures encompasses one or more subfields of machine learning and artificial intelligence. The primary goal of this symposium is to bring together practitioners in each of these areas and focus discussion on combining these lines of research toward lifelong machine learning.

Topics

The symposium will include paper presentations, talks, and discussions on a variety of topics related to lifelong learning, including but not limited to:

knowledge transfer

active transfer learning
multi-task learning
cross-domain transfer
knowledge/schema mapping
source knowledge selection
one-shot learning
transfer over long sequences of tasks

continual learning

online multi-task learning
online representation learning
knowledge maintenance/revision
developmental learning
scalable transfer learning
task/concept drift
self-selection of tasks

representation discovery

learning from raw sensory data
deep learning
latent representations
multi-modal/multi-view learning
multi-scale representations

incorporating guidance from external teachers

learning from demonstration
skill shaping
curriculum-based training
interactive learning
corrective feedback
agent-teacher communication

frameworks for lifelong learning

architectures
software frameworks
testbeds
evaluation methodology

applications of lifelong learning

data sets
application domains/environments
simulators
deployed applications

Within these topics, the symposium will explore lifelong learning in different problem formats, including classification, regression, and sequential decision-making problems.

Invited Speakers

Rich Sutton, University of Alberta
Jeff Dean, Google: "Large-Scale Learning from Multimodal Data"
Paul Ruvolo, Bryn Mawr College: "Efficient Lifelong Machine Learning"
Matthew Taylor, Washington State University: "Agents as Teachers and Learners"

Schedule

[PDF version of schedule]

Monday, March 25, 2013
9:00 - 9:15	Opening and Welcome Eric Eaton, Bryn Mawr College (Chair)
9:15 - 10:30	Invited Talk: Rich Sutton, U. Alberta
10:30 - 11:00	Coffee break
11:00 - 12:00	Paper Presentations
	Lifelong Machine Learning Systems: Beyond Learning Algorithms Daniel Silver, Qiang Yang & Lianghao Li Information-Theoretic Objective Functions for Lifelong Learning Byoung-Tak Zhang Towards Pareto Descent Directions in Sampling Experts for Multiple Tasks in an Online Learning Paradigm Shaona Ghosh, Chris Lovell & Steve Gunn
12:00 - 12:30	Discussion
12:30 - 2:00	Lunch break
2:00 - 3:10	Invited Talk: Jeff Dean, Google Large-Scale Learning from Multimodal Data
3:10 - 3:30	Paper Presentation
	Learning Sensorimotor Concepts Without Reinforcement Yasser Mohammad and Toyoaki Nishida
3:30 - 4:00	Coffee break
4:00 - 5:00	Paper Presentations
	Online Object Representation Learning and Its Application to Object Tracking Amirreza Shaban, Hamid R. Rabiee, Mehrdad Farajtabar & Mohsen Fadaee Organizing Behavior into Temporal and Spatial Neighborhoods Mark Ring & Tom Schaul Autonomous Selection of Inter-Task Mappings in Transfer Learning Anestis Fachantidis, Ioannis Partalas, Matthew E. Taylor & Ioannis Vlahavas
5:00 - 5:30	Discussion
5:30 - 6:00	Break
6:00 - 7:00	Reception

Tuesday, March 26, 2013
9:00 - 10:00	Invited Talk: Paul Ruvolo, Bryn Mawr Efficient Lifelong Machine Learning
10:00 - 10:30	Emerging Applications
	A Feedback-enabled Machine Learning Approach for Multi-Engine Machine Translation (Short Paper) Christian Federmann Applications of Lifelong Learning to the Google Knowledge Graph Terran Lane
10:30 - 11:00	Coffee break
11:00 - 12:00	Paper Presentations
	Lifelong Learning of Structure in the Space of Policies Majd Hawasly & Subramanian Ramamoorthy Automatic Abstraction in Reinforcement Learning Using Ant System Algorithm Nasrin Taghizadeh, Mohsen Ghafoorin & Hamid Beigy Hashing for Lightweight Episodic Recall Scott Wallace, Evan Dickinson & Andrew Nuxoll
12:00 - 12:30	Discussion
12:30 - 2:00	Lunch break
2:00 - 2:10	Paper Presentation
	The Consolidation of Task Knowledge for Lifelong Machine Learning (Short Paper) Daniel Silver
2:10 - 3:10	Invited Talk: Matthew Taylor, Washington State University Agents as Teachers and Learners
3:10 - 3:30	Organization of Working Sessions
3:30 - 4:00	Coffee break
4:00 - 5:30	Working Sessions
5:30 - 6:00	Break
6:00 - 7:00	Plenary Session

Wednesday, March 27, 2013
9:00 - 10:30	Reports by Working Session Leaders Discussion
10:30 - 11:00	Coffee break
11:00 - 12:30	Open discussion: Next steps toward lifelong machine learning

Important Dates

Please note that we have extended the submission deadline for papers in order to eliminate deadline conflicts with other AI-related conferences.

~~October 5, 2012~~ October 15, 2012 - Submissions due via EasyChair by 11:59:59pm Pacific Time
~~November 2, 2012~~ November 9, 2012 - Notification of acceptance/rejection sent to authors
January 18, 2013 - Final camera-ready papers and signed distribution license due to AAAI
February 15, 2013 - Invited participants registration deadline
March 8, 2013 - Final (open) registration deadline
March 25-27, 2013 - Symposium at Stanford University, California

Organizing Committee

Chair:
Eric Eaton (Bryn Mawr College, )

Members:
Terran Lane (Google, )
Honglak Lee (University of Michigan, )
Michael Littman (Brown University, )
Fei Sha (University of Southern California, )
Thomas Walsh (University of Kansas, )

Program Committee

Adam Coates (Stanford University)
Alan Fern (Oregon State University)
Quoc Le (Stanford University)
Clayton Morrison (University of Arizona)
Diane Oyen (University of New Mexico)
Paul Ruvolo (Bryn Mawr College)
Danny Silver (Acadia University)
Monica Vroman (Rutgers University)