Human Boosting

Harsh Pareek, Pradeep Ravikumar
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(1):338-346, 2013.

Abstract

Humans may be exceptional learners but they have biological limitations and moreover, inductive biases similar to machine learning algorithms. This puts limits on human learning ability and on the kinds of learning tasks humans can easily handle. In this paper, we consider the problem of “boosting” human learners to extend the learning ability of human learners and achieve improved performance on tasks which individual humans find difficult. We consider classification (category learning) tasks, propose a boosting algorithm for human learners and give theoretical justifications. We conduct experiments using Amazon’s Mechanical Turk on two synthetic datasets – a crosshair task with a nonlinear decision boundary and a gabor patch task with a linear boundary but which is inaccessible to human learners – and one real world dataset – the Opinion Spam detection task introduced in (Ott et al). Our results show that boosting human learners produces gains in accuracy and can overcome some fundamental limitations of human learners.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-pareek13, title = {Human Boosting}, author = {Pareek, Harsh and Ravikumar, Pradeep}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {338--346}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/pareek13.pdf}, url = {https://proceedings.mlr.press/v28/pareek13.html}, abstract = {Humans may be exceptional learners but they have biological limitations and moreover, inductive biases similar to machine learning algorithms. This puts limits on human learning ability and on the kinds of learning tasks humans can easily handle. In this paper, we consider the problem of “boosting” human learners to extend the learning ability of human learners and achieve improved performance on tasks which individual humans find difficult. We consider classification (category learning) tasks, propose a boosting algorithm for human learners and give theoretical justifications. We conduct experiments using Amazon’s Mechanical Turk on two synthetic datasets – a crosshair task with a nonlinear decision boundary and a gabor patch task with a linear boundary but which is inaccessible to human learners – and one real world dataset – the Opinion Spam detection task introduced in (Ott et al). Our results show that boosting human learners produces gains in accuracy and can overcome some fundamental limitations of human learners.} }
Endnote
%0 Conference Paper %T Human Boosting %A Harsh Pareek %A Pradeep Ravikumar %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-pareek13 %I PMLR %P 338--346 %U https://proceedings.mlr.press/v28/pareek13.html %V 28 %N 1 %X Humans may be exceptional learners but they have biological limitations and moreover, inductive biases similar to machine learning algorithms. This puts limits on human learning ability and on the kinds of learning tasks humans can easily handle. In this paper, we consider the problem of “boosting” human learners to extend the learning ability of human learners and achieve improved performance on tasks which individual humans find difficult. We consider classification (category learning) tasks, propose a boosting algorithm for human learners and give theoretical justifications. We conduct experiments using Amazon’s Mechanical Turk on two synthetic datasets – a crosshair task with a nonlinear decision boundary and a gabor patch task with a linear boundary but which is inaccessible to human learners – and one real world dataset – the Opinion Spam detection task introduced in (Ott et al). Our results show that boosting human learners produces gains in accuracy and can overcome some fundamental limitations of human learners.
RIS
TY - CPAPER TI - Human Boosting AU - Harsh Pareek AU - Pradeep Ravikumar BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-pareek13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 1 SP - 338 EP - 346 L1 - http://proceedings.mlr.press/v28/pareek13.pdf UR - https://proceedings.mlr.press/v28/pareek13.html AB - Humans may be exceptional learners but they have biological limitations and moreover, inductive biases similar to machine learning algorithms. This puts limits on human learning ability and on the kinds of learning tasks humans can easily handle. In this paper, we consider the problem of “boosting” human learners to extend the learning ability of human learners and achieve improved performance on tasks which individual humans find difficult. We consider classification (category learning) tasks, propose a boosting algorithm for human learners and give theoretical justifications. We conduct experiments using Amazon’s Mechanical Turk on two synthetic datasets – a crosshair task with a nonlinear decision boundary and a gabor patch task with a linear boundary but which is inaccessible to human learners – and one real world dataset – the Opinion Spam detection task introduced in (Ott et al). Our results show that boosting human learners produces gains in accuracy and can overcome some fundamental limitations of human learners. ER -
APA
Pareek, H. & Ravikumar, P.. (2013). Human Boosting. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(1):338-346 Available from https://proceedings.mlr.press/v28/pareek13.html.

Related Material