Research Overview
How does the biological mind work? How does the brain build
a model of the world to instantly and robustly act on it? What kinds of
processing and organizational mechanisms allow the brain to learn so rapidly,
flexibly, and continuously in a noisy, dynamic, and changing environment? How
can we build smart machines that perceive, think, act, and learn like the
human brain and mind? In the Biointelligence Lab and the Cognitive Robotics
and ArtificialIntelligence Center (CRAIC) we pursue these questions using
diverse methods of cognitive modeling, computational neuroscience, robotics,
and computer science.
We
build robotic surrogates by "cloning" the embodied and situated
mind using wearable sensors, such as smart glasses, watches, brain scanners,
body sensors, and mobile devices. We collect human activity data by long-term
life-tracking in the real world. Based on these observational data, we
reverse-engineer the brain architecture by exploring the plausible
neurocognitive models that best explain the mind in working. We also use
videos and cartoons to study how the cognitive brain constructs episodic
memory and mental imagery from a long sequence of context-rich events while
watching dramas and movies.
Our
current efforts center around the deep, recurrent, and sparse hypernetwork
architectures and learning algorithms that self-organize their structures instantly,
incrementally, and continuously in a self-supervised way by perception-action
cycle. The ultimate goal is to discover a large-scale, neurocognitive
computational model of the brain that autonomously develops or evolves into
human-level machine intelligence in lifelong interactions with the
environment. For the ongoing and completed projects, visit BI
research. For general information, visit the Biointelligence Laboratory home.
Theory:
Hypernetwork Models of Learning and Memory
Applications:
Data Mining in Life Sciences and Life Blogs
Technology:
Molecular Evolutionary Computation in Silico and in Vitro
Talks and Tutorials_International
Talks and Tutorials_Domestic
Position Papers & Monographs [Korean]
Meetings Organized
Selected Publications
-
Attend What You Need: Motion-Appearance Synergistic Networks for Video Question Answering,
A.-J. Seo, G.-C. Kang, J.-H. Park, B.-T. Zhang,
Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021),
August 2021.
-
Message Passing Adaptive Resonance Theory for Online Active Semi-supervised Learning,
T. Kim, I. Hwang, H.-D. Lee, H. Kim, W.-S. Choi, J. Lim, B.-T. Zhang,
Proceedings of the 38th International Conference on Machine Learning (ICML 2021),
July 2021.
-
Co-attentional Transformers for Story-Based Video Understanding,
B. Bebensee, B.-T. Zhang,
Proceedings of the 46th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2020),
June 2021.
[PDF]
-
DramaQA: Character-Centered Video Story Understanding with Hierarchical QA,
S.-H. Choi, K.-W. On, Y.-J Heo, A.-J. Seo, Y.-W. Jang, M.-S. Lee and B.-T. Zhang,
Proceedings of the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI 2021),
February 2021.
[PDF]
-
Hypergraph Attention Networks for Multimodal Learning,
E.-S. Kim, W.-Y. Kang, K.-W. On, Y.-J. Heo and B.-T. Zhang,
The IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 2020),
2020.
[PDF]
-
Cut-Based Graph Learning networks to Discover Compositional Structure of Sequential Video Data,
K.-W. On, E.-S. Kim, Y.-J Heo and B.-T. Zhang,
Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020),
2020.(oral presentation, accept ratio=5.8%)
[PDF]
-
Answerer in questioner's mind: Information theoretic approach to goal-oriented visual dialog,
S.-W. Lee, Y.-J. Heo, B.-T. Zhang,
The 32nd Annual Conference on Neural Information Processing Systems (NIPS 2018),
2018.
(Spotlight)
[PDF]
-
Bilinear attention networks,
J.-H. Kim, J. Jun, B.-T. Zhang,
The 32nd Annual Conference on Neural Information Processing Systems (NIPS 2018),
2018.
[PDF]
-
Multimodal dual attention memory for video story question answering,
K. Kim, S.-H. Choi, J.-H. Kim, B.-T. Zhang,
The 15th European Conference on Computer Vision (ECCV 2018),
2018.
[PDF]
-
Robust human following by deep Bayesian trajectory prediction for home service robots,
B.-J. Lee, J. Choi, C. Baek, B.-T. Zhang,
2018 IEEE International Conference on Robotics and Automation (ICRA 2018),
2018.
[PDF]
-
Human Intelligence and Machine Intelligence: Cognitive Artificial Intelligence (in Korean),
Zhang, B.-T.,
Communications of the KIISE,
36(1): 27-36,
2018.
[PDF]
- Overcoming catastrophic forgetting by incremental moment matching, S.-W. Lee, J.-H. Kim, J. Jun, J.-W. Ha, and B.-T. Zhang, The 31st Annual Conference on Neural Information Processing Systems (NIPS 2017),
2017. (Spotlight) [PDF]
- Dual-memory neural networks for modeling cognitive activities of humans via wearable sensors, S.-W. Lee, C.-Y. Lee, D.-H. Kwak, J.-W. Ha, J. Kim, and B.-T. Zhang, Neural Networks
,
92:17-28,
2017. [PDF]
- DeepStory: Video story QA by deep embedded
memory networks, K.-M. Kim, M.-O. Heo, S.-H. Choi, and B.-T. Zhang, The 26th International Joint
Conference on Artificial Intelligence (IJCAI 2017),
2017. [PDF]
- Hadamard product for low-rank bilinear pooling,
J.-H. Kim, K.-W. On, W. Lim, J. Kim, J.-W. Ha, B.-T. Zhang, The 5th International
Conference on Learning Representations (ICLR 2017),
2017. [PDF]
- Humans and machines in the evolution of AI in
Korea, AI
Magazine, 37(2):108-112, 2016. [PDF]
- Multimodal residual learning for visual QA,
J.-H. Kim, S.-W. Lee, D.-H. Kwak, M.-O. Heo, J. Kim, J.-W. Ha, and
B.-T. Zhang, In The 30th Annual Conference on Neural Information Processing Systems (NIPS
2016) [PDF]
- DeepSchema: Automatic schema acquisition from
wearable sensor data in restaurant situations, E.-S. Kim, K.-W. On,
and B.-T. Zhang, In Proceedings of the International Joint
Conference on Artificial Intelligence (IJCAI 2016), pp.
834-840, 2016. [PDF]
- Dual-memory deep learning architectures for
lifelong learning of everyday human behaviors, S.-W. Lee, C.-Y. Lee,
D. H. Kwak, J. Kim, J. Kim, and B.-T. Zhang, In Proceedings
of the International Joint Conference on Artificial Intelligence
(IJCAI 2016), 1669-1675, 2016. [PDF]
- Deep hypernetwork models (in Korean), Communications of the Korean
Institute of Information Scientists and Engineers,
33(8):11-24, 2015. [PDF]
- Team THOR's entry in the DARPA robotics
challenge trials 2013, Journal of Field Robotics,
32(3):315-335, 2015. [PDF]
- Consensus analysis and modeling of visual
aesthetic perception, IEEE Transactions on Affective Computing,
6(3):272-285. [PDF]
- Automated construction of visual-linguistic
knowledge via concept learning from cartoon videos, J.-W. Ha, K.-M.
Kim, and B.-T. Zhang, In Proceedings of the Twenty-Ninth
AAAI Conference on Artificial Intelligence (AAAI 2015), 2015.
[PDF]
- Ontogenesis of agency in machines: A
multidisciplinary review, B.-T. Zhang, AAAI 2014 Fall Symposium
on The Nature of Humans and Machines: A Multidisciplinary Discourse,
Arlington, VA, 2014. [PDF]
- Communication as moving target tracking: Dynamic
Bayesian inference with an action-perception-learning cycle, B.-T.
Zhang, Alignment in
Communication. Towards a New Theory of Communication, Chapter
7, pp. 133-147, 2013. [PDF]
- Information-theoretic objective functions for
lifelong learning, B.-T. Zhang, AAAI
2013 Spring Symposium on Lifelong Machine Learning,
Stanford University, March 25-27, 2013. [PDF]
- Sparse population code models of word learningin
concept drift, B.-T. Zhang, J.-W. Ha, and M. Kang, In Proceedings of the 34th Annual Meeting of the Cognitive
Science Society (CogSci 2012), 2012.
[PDF]
- A DNA assembly model of
sentence generation, J.-H. Lee, S. H. Lee, W.-H. Chung, E. S. Lee, T.
H. Park, R. Deaton, and B.-T. Zhang, BioSystems,
106:51-55, 2011. [PDF]
- Hypernetworks:
A molecular evolutionary architecture for cognitive learning and
memory, B.-T. Zhang, IEEE Computational Intelligence Magazine,
3(3):49-63, August 2008.[PDF][LINK]
- Molecular basis for the
recognition of primary microRNAs by the Drosha-DGCR8 complex, J. Han, Lee
Y, K.H. Yeom, J.-W. Nam, I. Heo, J.-K. Rhee, S. Y. Sohn, Y. Cho,
B.-T. Zhang and V.N. Kim, Cell,
125:887-901, 2006. [PDF]
- Multi-objective
evolutionary optimization of DNA sequences for reliable DNA computing,
S.-Y. Shin, I.-H. Lee, D. Kim, and B.-T. Zhang, IEEE Transactions on Evolutionary Computation,
9(2):143-158, 2005. [PDF]
- Bayesian
model averaging of Bayesian network classifiers over multiple
node-orders: application to sparse datasets, K.-B. Hwang and B.-T.
Zhang, IEEE Transactions on
Systems, Man, and Cybernetics Part B: Cybernetics,
35(6):1302-1310, 2005. [PDF]
- Self-organizing
latent lattice models for temporal gene expression profiling, Zhang,
B.-T, Yang, J. and Chi, S. W., Machine Learning,
52(1/2):67-89, 2003. [PDF]
- System
identification using evolutionary Markov chain monte carlo, Zhang,
B.-T. and Cho, D.-Y., Journal of Systems Architecture,
47(7):587-599, 2001. [PDF]
- Bayesian
methods for efficient genetic programming, Zhang, B.-T., Genetic
Programming and Evolvable Machines, 1(3):217-242, 2000. [PDF]
- Evolutionary
induction of sparse neural trees, Zhang, B.-T., Ohm, P., and
M?hlenbein, H., Evolutionary Computation, 5(2):213-236,
1997. [PDF]
- Balancing
accuracy and parsimony in genetic programming, Zhang, B.-T. and
M?hlenbein, H., Evolutionary Computation, 3(1):17-38,
1995. [PDF]
- Accelerated
learning by active example selection, Zhang, B.-T., International
Journal of Neural Systems, 5(1):67-75, 1994. [PDF]
- Evolving
optimal neural networks using genetic algorithms with Occam's razor,
Zhang, B.-T. and M?hlenbein, H., Complex Systems,
7(3):199-220, 1993. [PDF] [LINK]
[more]
|
|