IUI workshop on interactive machine learning

Published 19 March 2013

Saleema Amershi, Maya Çakmak, W. Bradley Knox, Todd Kulesza, Tessa Lau

Citations4

TL;DR

This Interactive Machine Learning workshop at IUI 2013 aims to bring this community of researchers at the intersection of ML and human-computer interaction together to share ideas, get up-to-date on recent advances, progress towards a common framework and terminology for the field, and discuss the open questions and challenges of IML.

Abstract

Many applications of Machine Learning (ML) involve interactions with humans. Humans may provide input to a learning algorithm (in the form of labels, demonstrations, corrections, rankings or evaluations) while observing its outputs (in the form of feedback, predictions or executions). Although humans are an integral part of the learning process, traditional ML systems used in these applications are agnostic to the fact that inputs/outputs are from/for humans.

Keywords

Computer Science

Active Learning

1,150 Citations2012Burr Settles

Synthesis lectures on artificial intelligence and machine learningActive Learning

446 Citations2012Burr Settles

Active learning by labeling features

160 Citations2009Gregory Druck, Burr Settles +1 more

This paper proposes an active learning approach in which the machine solicits "labels" on features rather than instances, and shows that this method outperforms passive learning with features as well as traditional active learning with instances.

View PDF

IEEE Transactions on Autonomous Mental DevelopmentDesigning Interactions for Robot Active Learners

148 Citations2010Maya Çakmak, Crystal Chao +1 more

Three interaction modes that enable a robot to use active learning queries are presented that are preferable to the mode using passive supervised learning both in terms of performance and human subject preference, but each mode has advantages and disadvantages.

Interactive optimization for steering machine classification

134 Citations2010Ashish Kapoor, Bongshin Lee +2 more

ManiMatrix is presented, a system that provides controls and visualizations that enable system builders to refine the behavior of classification systems in an intuitive manner and results show that users are able to quickly and effectively modify decision boundaries of classifiers to tai-lor thebehavior of classifier to problems at hand.

Reinforcement learning from human reward: Discounting in episodic tasks

67 Citations2012W. Bradley Knox, Peter Stone

Using model-based reinforcement learning from human reward in goal-based, episodic tasks, this work investigates how anticipated future rewards should be discounted to create behavior that performs well on the task that the human trainer intends to teach.

Integrating rich user feedback into intelligent user interfaces

45 Citations2008Simone Stumpf, Erin Sullivan +4 more

It was found that rich feedback improved accuracy but an initial unstable period often caused large fluctuations in classifier behavior and showed that in order to benefit from the user's knowledge, machine learning systems must be able to absorb keyword-based rich feedback in a graceful manner and provide clear explanations of their predictions.

CueT

44 Citations2011Saleema Amershi, Bongshin Lee +3 more

CueT is presented, a system that uses interactive machine learning to learn from the triaging decisions of operators and then uses that learning in novel visualizations to help operators quickly and accurately triage alarms.

Towards maximizing the accuracy of human-labeled sensor data

35 Citations2010Stephanie Rosenthal, Anind K. Dey

Two studies that evaluate the accuracy of human responses to an intelligent agent's data classification questions find very similar combinations of information (for users and strangers) that led to very accurate responses as well as more feedback that the agents could use to refine their predictions.