Can Active Learning Experience Be Transferred?

Active learning is an important machine learning problem in reducing the human labeling effort. Current active learning strategies are designed from human knowledge, and are applied on each dataset in an immutable manner. In other words, experience about the usefulness of strategies cannot be updated and transferred to improve active learning on other datasets. This paper initiates a pioneering study on whether active learning experience can be transferred. We first propose a novel active learning model that linearly aggregates existing strategies. The linear weights can then be used to represent the active learning experience. We equip the model with the popular linear upper-confidence-bound (LinUCB) algorithm for contextual bandit to update the weights. Finally, we extend our model to transfer the experience across datasets with the technique of biased regularization. Empirical studies demonstrate that the learned experience not only is competitive with existing strategies on most single datasets, but also can be transferred across datasets to improve the performance on future learning tasks.

[1]  Ying Liu,et al.  Active Learning with Support Vector Machine Applied to Gene Expression Data for Cancer Classification , 2004, J. Chem. Inf. Model..

[2]  Yan Liu,et al.  Hierarchical Active Transfer Learning , 2015, SDM.

[3]  Wei Chu,et al.  Contextual Bandits with Linear Payoff Functions , 2011, AISTATS.

[4]  Xinlei Chen,et al.  Never-Ending Learning , 2012, ECAI.

[5]  Peter Auer,et al.  Using Confidence Bounds for Exploitation-Exploration Trade-offs , 2003, J. Mach. Learn. Res..

[6]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[7]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[8]  Xiaowei Xu,et al.  Representative Sampling for Text Classification Using Support Vector Machines , 2003, ECIR.

[9]  Wei Chu,et al.  A contextual-bandit approach to personalized news article recommendation , 2010, WWW '10.

[10]  Wei Fan,et al.  Actively Transfer Domain Knowledge , 2008, ECML/PKDD.

[11]  John Langford,et al.  Contextual Bandit Algorithms with Supervised Learning Guarantees , 2010, AISTATS.

[12]  Eric Eaton,et al.  ELLA: An Efficient Lifelong Learning Algorithm , 2013, ICML.

[13]  William A. Gale,et al.  A sequential algorithm for training text classifiers , 1994, SIGIR '94.

[14]  Yan Liu,et al.  Accelerating Active Learning with Transfer Learning , 2013, 2013 IEEE 13th International Conference on Data Mining.

[15]  Burr Settles,et al.  Active Learning Literature Survey , 2009 .

[16]  Tsuhan Chen,et al.  An active learning framework for content-based information retrieval , 2002, IEEE Trans. Multim..

[17]  Kristen Grauman,et al.  Learning with Whom to Share in Multi-task Feature Learning , 2011, ICML.

[18]  Chun-Liang Li,et al.  Active Learning Using Hint Information , 2015, Neural Computation.

[19]  Peter Auer,et al.  The Nonstochastic Multiarmed Bandit Problem , 2002, SIAM J. Comput..

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

[21]  Hsuan-Tien Lin,et al.  Active Learning by Learning , 2015, AAAI.

[22]  Bing Liu,et al.  Lifelong Learning for Sentiment Classification , 2015, ACL.

[23]  Kumar Chellapilla,et al.  Personalized handwriting recognition via biased regularization , 2006, ICML.

[24]  Daphne Koller,et al.  Support Vector Machine Active Learning with Applications to Text Classification , 2000, J. Mach. Learn. Res..

[25]  Shipra Agrawal,et al.  Thompson Sampling for Contextual Bandits with Linear Payoffs , 2012, ICML.

[26]  Paul N. Bennett,et al.  Dual Strategy Active Learning , 2007, ECML.

[27]  Ran El-Yaniv,et al.  Online Choice of Active Learning Algorithms , 2003, J. Mach. Learn. Res..

[28]  Rong Jin,et al.  Active Learning by Querying Informative and Representative Examples , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.