ELLA: An Efficient Lifelong Learning Algorithm

The problem of learning multiple consecutive tasks, known as lifelong learning, is of great importance to the creation of intelligent, general-purpose, and flexible machines. In this paper, we develop a method for online multi-task learning in the lifelong learning setting. The proposed Efficient Lifelong Learning Algorithm (ELLA) maintains a sparsely shared basis for all task models, transfers knowledge from the basis to learn each new task, and refines the basis over time to maximize performance across all tasks. We show that ELLA has strong connections to both online dictionary learning for sparse coding and state-of-the-art batch multitask learning methods, and provide robust theoretical performance guarantees. We show empirically that ELLA yields nearly identical performance to batch multi-task learning while learning tasks sequentially in three orders of magnitude (over 1,000x) less time.

[1]  J. Danskin.  The Theory of Max-Min and its Application to Weapons Allocation Problems , 1967 .

[2]  Kai-Bor Yu,et al.  Recursive updating the eigenvalue decomposition of a covariance matrix , 1991, IEEE Trans. Signal Process..

[3]  Sebastian Thrun,et al.  Explanation-based neural network learning a lifelong learning approach , 1995 .

[4]  Sebastian Thrun,et al.  Discovering Structure in Multiple Learning Tasks: The TC Algorithm , 1996, ICML.

[5]  Alexander Shapiro,et al.  Optimization Problems with Perturbations: A Guided Tour , 1998, SIAM Rev..

[6]  Mark B. Ring.  CHILD: A First Step Towards Continual Learning , 1997, Machine Learning.

[7]  Lawrence Carin,et al.  Multi-Task Learning for Classification with Dirichlet Process Priors , 2007, J. Mach. Learn. Res..

[8]  Richard S. Sutton,et al.  On the role of tracking in stationary environments , 2007, ICML '07.

[9]  Yiming Yang,et al.  Flexible latent variable models for multi-task learning , 2008, Machine Learning.

[10]  Guillermo Sapiro,et al.  Online dictionary learning for sparse coding , 2009, ICML '09.

[11]  Estevam R. Hruschka,et al.  Toward an Architecture for Never-Ending Language Learning , 2010, AAAI.

[12]  Hal Daumé,et al.  Infinite Predictor Subspace Models for Multitask Learning , 2010, AISTATS.

[13]  Avishek Saha,et al.  Online Learning of Multiple Tasks and Their Relationships , 2011, AISTATS.

[14]  Masashi Sugiyama,et al.  Computationally Efficient Multi-task Learning with Least-squares Probabilistic Classifiers , 2010, IPSJ Trans. Comput. Vis. Appl..

[15]  Maja Pantic,et al.  The first facial expression recognition and analysis challenge , 2011, Face and Gesture 2011.

[16]  Hal Daumé,et al.  Learning Task Grouping and Overlap in Multi-task Learning , 2012, ICML.