Prediction is very difficult, especially about the future.
-- Niels Bohr
Affective Computing Group
at the MIT Media Lab
PhD in Media Arts and Sciences, MIT
PhD dissertation (2010) title: Modeling and Analysis of Affective Influences on Human Experience, Prediction, Decision Making, and Behavior
Advisor : Rosalind W. Picard
Computational Models of Human Behavioral, Affective and Cognitive Decision Making
Machine Learning, Human-Agent Interaction, Computational User Experience, Computational Customer Experience
Prediction Models of Customer Decisions, Prediction Market, Agent-based Computational Economics
Neuroeconomics, Behavioral Economics, Consumer Psychology
Affective Cognitive Learning and Decision Making
Recent affective neuroscience and psychology indicate that human affect and emotional experience play a significant, and useful, role in human learning and decision making. Most machine learning and decision-making models, however, are based on old purely cognitive models. I aim to redress this problem, by developing new models that integrate affect with cognition. The first model is being built to address several very difficult problems in machine learning. My aim is to utilize affect-like mechanisms to fix several of these problems. I expect that an integrated affective-cognitive learning system should exhibit many improvements over the state of the art, ultimately enabling much smoother human-computer interaction and more intelligent human-machine systems.
A Multi-Modal Affective-Cognitive Technique for Product Evaluation with Computational Models of Predicting Customer Decisions
Companies want to get more new products to be successful in the marketplace; however, current evaluation methods such as focus groups do not accurately predict customer decisions in the marketplace. We are developing new technology-assisted methods to try to improve the customer evaluation process and better predict customer decisions. The new methods involve multi-modal affective measures (such as facial expression and skin conductance) together with behavioral measures, anticipatory-motivational measures, and self-report cognitive measures. These measures are combined into a novel computational model, the form of which is motivated by findings in affective neuroscience and human behavior. The model is being trained and tested with customer product evaluations and marketplace outcomes from real product launches.
The early stage of this research was covered on The Wall Street Journal, "MIT Researchers Read Consumers' Faces to Make a Better Taste Test," January 19, 2010. (click here to see the article)
The US patent related to this research was filed on 09/25/2009. U.S. Patent Application 12/567115 "Multimodal Affective-Cognitive Product Evaluation," Hyungil Ahn and Rosalind Picard.
Prediction Game and Experience Sharing Market for Forecasting Marketplace Success
We have developed a novel market game, "Prediction Game and Experience Sharing" (PreGES, pronounced as PreGuess), that harnesses people's collective prediction and experience sharing to forecast the success or failure of new items (e.g., products, services, UI designs). Companies can register their new items on this market (as a testbed) to ask people's collective opinion. In each PreGES trial session, a participant makes his or her own best prediction on other people's overall opinions about the new items to get incentives (e.g., real opportunities to experience the items) and have fun in gambling-like games. As a participant's guess (or portfolio) approaches the collective guess of all participants, he or she has a greater chance of winning an incentive. Participants improve the accuracy of their next prediction by sharing experiences. As participants have more trial sessions, their collective prediction converges into one common opinion (forecasting the success or failure of new items).
Robotic Computer (RoCo)
A robotic computer that moves its monitor "head" and "neck,"
but that has no explicit face, is being designed to interact with users
in a natural way for applications such as learning, rapport-building,
interactive teaching, and posture improvement. In all these applications,
the robot will need to move in subtle ways that express its state and
promote appropriate movements in the user, but that don't distract or
annoy. Toward this goal, we are giving the system the ability to recognize
and have subtle expressions. This project is an ongoing collaboration
with the MIT Media Lab's Personal Robots
To see more details, please refer to
Hyungil Ahn, Alea Teeters, Andrew Wang, Cynthia Breazeal and Rosalind Picard (2007), "Stoop to Conquer: Posture and affect interact to influence computer users' persistence" (pdf)
NewScientist article on RoCo, March 22, 2007: If you're happy, the robot knows it (click here to see the article)
The U.S. patent related to this research was published on 12/24/2009. U.S. Patent Application 20090319459, "Physically-animated Visual Display," Cynthia Breazeal, Rosalind Picard, Hyungil Ahn and Guy Hoffman
(MAS.750) Human-Robot Interaction, Cynthia Breazeal, Fall 2006
(MAS.921) PhD Proseminar in Media Arts and Sciences, Deb Roy, Fall 2006
Hyung-il Ahn, Werner Geyer, Casey Dugan, David Millen (2009), "How incredibly awesome! - click hear to read more," (pdf) International AAAI Conference on Weblogs and Social Media (ICWSM 2010), May 2010, Washington, DC, USA.
Hyung-il Ahn and Rosalind Picard (2009), "Affective-Cognitive Prediction Framework for Customer Preferences and Decisions," Working Paper
Hyung-il Ahn and Rosalind Picard (2009), "Affective-Subjective Prediction using Prospect Theory in a Reinforcement Learning Framework," Working Paper
Hyung-il Ahn and Dustin Smith (2009), "Action Planning with Commonsense Knowledge," (pdf) Work-in-progress in the ACM Conference on Human Factors in Computing Systems (CHI 2009), April 2009, Boston, MA, USA.
Hyung-il Ahn, Alea Teeters, Andrew Wang, Cynthia Breazeal and Rosalind Picard (2007), "Stoop to Conquer: Posture and affect interact to influence computer users' persistence" (pdf), The 2nd International Conference on Affective Computing and Intelligent Interaction (ACII 2007), September 12-14, 2007, Lisbon, Portugal.
Hyung-il Ahn and Rosalind W. Picard (2006), "Affective-Cognitive Learning and Decision Making: The Role of Emotions" (pdf), The 18th European Meeting on Cybernetics and Systems Research (EMCSR 2006). April 18-19, 2006, Vienna, Austria.
J.J. Goo, K.S. Park, M. Lee, J. Park, M. Hahn, H. Ahn, and R. W. Picard (2006), "Effects of Guided and Unguided Style Learning on User Attention in a Virtual Environment" (pdf), Edutainment2006.
Ashish Kapoor, Yuan (Alan) Qi, Hyung-il Ahn and Rosalind W. Picard (2005),
"Hyperparameter and Kernel Learning for Graph Based Semi-Supervised
Classification" (pdf), Neural
Information Processing System (NIPS 2005). December 5-8, 2005, Vancouver,
Hyung-il Ahn and Rosalind W. Picard (2005), "Affective-Cognitive Learning and Decision Making: A Motivational Reward Framework For Affective Agents" (pdf), The 1st International Conference on Affective Computing and Intelligent Interaction (ACII 2005). October 22-24, 2005, Beijing, China.
Ashish Kapoor, Hyungil Ahn and Rosalind W. Picard (2005), "Mixture
of Gaussian Processes for Combining Multiple Modalities" (pdf),
Workshop on Multiple Classifier Systems (MCS 2005).
U.S. Patent Application 12/567115 "Multimodal Affective-Cognitive Product Evaluation," Hyungil Ahn and Rosalind Picard, filing date: 09/25/2009.
U.S. Patent Application 20090319459, "Physically-animated Visual Display," Cynthia Breazeal, Rosalind Picard, Hyungil Ahn and Guy Hoffman, publication date: 12/24/2009.
MIT Media Laboratory
75 Amherst Street, E14-374H
Cambridge, MA 02139
(Email) hiahn aT media doT mit doT edu