Active Transfer Learning under Model Shift

Xuezhi Wang, Tzu-Kuo Huang, Jeff Schneider
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):1305-1313, 2014.

Abstract

Transfer learning algorithms are used when one has sufficient training data for one supervised learning task (the source task) but only very limited training data for a second task (the target task) that is similar but not identical to the first. These algorithms use varying assumptions about the similarity between the tasks to carry information from the source to the target task. Common assumptions are that only certain specific marginal or conditional distributions have changed while all else remains the same. Alternatively, if one has only the target task, but also has the ability to choose a limited amount of additional training data to collect, then active learning algorithms are used to make choices which will most improve performance on the target task. These algorithms may be combined into active transfer learning, but previous efforts have had to apply the two methods in sequence or use restrictive transfer assumptions. We propose two transfer learning algorithms that allow changes in all marginal and conditional distributions but assume the changes are smooth in order to achieve transfer between the tasks. We then propose an active learning algorithm for the second method that yields a combined active transfer learning algorithm. We demonstrate the algorithms on synthetic functions and a real-world task on estimating the yield of vineyards from images of the grapes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-wangi14, title = {Active Transfer Learning under Model Shift}, author = {Wang, Xuezhi and Huang, Tzu-Kuo and Schneider, Jeff}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {1305--1313}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/wangi14.pdf}, url = {https://proceedings.mlr.press/v32/wangi14.html}, abstract = {Transfer learning algorithms are used when one has sufficient training data for one supervised learning task (the source task) but only very limited training data for a second task (the target task) that is similar but not identical to the first. These algorithms use varying assumptions about the similarity between the tasks to carry information from the source to the target task. Common assumptions are that only certain specific marginal or conditional distributions have changed while all else remains the same. Alternatively, if one has only the target task, but also has the ability to choose a limited amount of additional training data to collect, then active learning algorithms are used to make choices which will most improve performance on the target task. These algorithms may be combined into active transfer learning, but previous efforts have had to apply the two methods in sequence or use restrictive transfer assumptions. We propose two transfer learning algorithms that allow changes in all marginal and conditional distributions but assume the changes are smooth in order to achieve transfer between the tasks. We then propose an active learning algorithm for the second method that yields a combined active transfer learning algorithm. We demonstrate the algorithms on synthetic functions and a real-world task on estimating the yield of vineyards from images of the grapes.} }
Endnote
%0 Conference Paper %T Active Transfer Learning under Model Shift %A Xuezhi Wang %A Tzu-Kuo Huang %A Jeff Schneider %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-wangi14 %I PMLR %P 1305--1313 %U https://proceedings.mlr.press/v32/wangi14.html %V 32 %N 2 %X Transfer learning algorithms are used when one has sufficient training data for one supervised learning task (the source task) but only very limited training data for a second task (the target task) that is similar but not identical to the first. These algorithms use varying assumptions about the similarity between the tasks to carry information from the source to the target task. Common assumptions are that only certain specific marginal or conditional distributions have changed while all else remains the same. Alternatively, if one has only the target task, but also has the ability to choose a limited amount of additional training data to collect, then active learning algorithms are used to make choices which will most improve performance on the target task. These algorithms may be combined into active transfer learning, but previous efforts have had to apply the two methods in sequence or use restrictive transfer assumptions. We propose two transfer learning algorithms that allow changes in all marginal and conditional distributions but assume the changes are smooth in order to achieve transfer between the tasks. We then propose an active learning algorithm for the second method that yields a combined active transfer learning algorithm. We demonstrate the algorithms on synthetic functions and a real-world task on estimating the yield of vineyards from images of the grapes.
RIS
TY - CPAPER TI - Active Transfer Learning under Model Shift AU - Xuezhi Wang AU - Tzu-Kuo Huang AU - Jeff Schneider BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-wangi14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 1305 EP - 1313 L1 - http://proceedings.mlr.press/v32/wangi14.pdf UR - https://proceedings.mlr.press/v32/wangi14.html AB - Transfer learning algorithms are used when one has sufficient training data for one supervised learning task (the source task) but only very limited training data for a second task (the target task) that is similar but not identical to the first. These algorithms use varying assumptions about the similarity between the tasks to carry information from the source to the target task. Common assumptions are that only certain specific marginal or conditional distributions have changed while all else remains the same. Alternatively, if one has only the target task, but also has the ability to choose a limited amount of additional training data to collect, then active learning algorithms are used to make choices which will most improve performance on the target task. These algorithms may be combined into active transfer learning, but previous efforts have had to apply the two methods in sequence or use restrictive transfer assumptions. We propose two transfer learning algorithms that allow changes in all marginal and conditional distributions but assume the changes are smooth in order to achieve transfer between the tasks. We then propose an active learning algorithm for the second method that yields a combined active transfer learning algorithm. We demonstrate the algorithms on synthetic functions and a real-world task on estimating the yield of vineyards from images of the grapes. ER -
APA
Wang, X., Huang, T. & Schneider, J.. (2014). Active Transfer Learning under Model Shift. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):1305-1313 Available from https://proceedings.mlr.press/v32/wangi14.html.

Related Material