Article
Psychology, Biological
Carolina Feher da Silva, Gaia Lombardi, Micah Edelson, Todd A. Hare
Summary: A standard assumption in neuroscience is challenged, as evidence suggests that model-based learning may not be automatic in humans.
NATURE HUMAN BEHAVIOUR
(2023)
Article
Education & Educational Research
Mahmoud Abdi Tabari, Xiaofei Lu, Yizhou Wang
Summary: This study investigated the effects of task complexity on the lexical complexity of L2 learners' written production. Results showed significantly higher levels of lexical complexity in the complex task compared to the simple task. The correlations between different word categories varied, with stronger correlations between total words and content words.
Article
Engineering, Electrical & Electronic
Geon Choi, Jeonghun Park, Nir Shlezinger, Yonina C. Eldar, Namyoon Lee
Summary: This paper introduces a robust EKF algorithm called Split-KalmanNet that utilizes deep learning for state estimation. Split-KalmanNet calculates the Kalman gain using a split structure and is able to compensate for state and measurement model mismatch effects, outperforming traditional EKF and the state-of-the-art KalmanNet algorithm in various scenarios of model mismatch.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Engineering, Civil
Mei-En Shao, Muhamad Amirul Haq, De-Qin Gao, Peter Chondro, Shanq-Jang Ruan
Summary: This research aims to simplify the complex learning process for driving scene understanding by proposing a method based on point of interest detection, training a single deep convolution neural network, supplemented with CV-based post-processing, for detection of free space and lane. Through verification on two publicly available datasets, achieved high accuracy.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Mei-En Shao, Muhamad Amirul Haq, De-Qin Gao, Peter Chondro, Shanq-Jang Ruan
Summary: This research aims to simplify the complex learning process of pixel-wise approach for driving scene understanding by implementing point detection for free space and lane detection. By training a single deep convolutional neural network for point of interest detection and using computer vision-based post-processing, the network achieved good results for pixel-wise detection of semantic segmentation and parametric description of lanes.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Psychiatry
Zhongqiang Ruan, Carol A. Seger, Qiong Yang, Dongjae Kim, Sang Wan Lee, Qi Chen, Ziwen Peng
Summary: Impaired arbitration mechanism for flexible adaptation to environmental demands in both OCD patients and healthy individuals reporting high OCI-R scores.
FRONTIERS IN PSYCHIATRY
(2023)
Article
Behavioral Sciences
Raul Luna, Miguel A. Vadillo, David Luque
Summary: Human behavior is supported by two computational learning mechanisms: model-free and model-based. Previous research suggests that participants' behavior usually involves a mixture of both strategies, but a recent study found that detailed instructions can increase the use of model-based behavior. This study replicated that experiment and confirmed that improved instructions enhance the model-based component. Additionally, the study found that the effect of reward is related to stimulus-response learning.
Article
Computer Science, Artificial Intelligence
Jingyao Zhang, Deyuan Meng
Summary: This article proposes a new design method to improve the tracking accuracy of iterative learning control (ILC) by adopting a high-order extended state observer (ESO). The designed ESO-based ILC achieves robust tracking of any desired trajectory and allows for regulation of the ILC tracking accuracy through the design of the ESO.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Florian Arnold, Rudibert King
Summary: This paper introduces a state-space model formulation using physics-informed neural networks to efficiently model complex dynamic systems. The resulting model does not require numerical solution techniques during state propagation and is suitable for real-time applications.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Review
Computer Science, Artificial Intelligence
Khimya Khetarpal, Matthew Riemer, Irina Rish, Doina Precup
Summary: This article provides a literature review of different formulations and approaches to continual reinforcement learning, discussing the perspective on why RL is suitable for studying continual learning, providing a taxonomy of different formulations and approaches, and discussing evaluation and future challenges.
JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH
(2022)
Article
Engineering, Aerospace
Lei Yang, Jue Huang, Qi Gao, Yi Zhou, Minghua Hu, Hua Xie
Summary: This paper investigates the issue of air traffic control workload in the Free Route Airspace (FRA). A complexity indicator system is constructed and the XGBoost algorithm is used to predict the workload. A two-stage sector boundary optimization method is proposed to achieve the goal of balancing the workload.
Article
Environmental Studies
Tao Cui, J. Sreekanth, Trevor Pickett, David Rassam, Mat Gilfedder, Damian Barrett
Summary: The parameterization strategy significantly influences the outputs and associated uncertainty of models, with simpler parameterization leading to wider prediction ranges and more complex parameterization providing more accurate results. The choice of model parameterization has a significant impact on predictive uncertainty and should be explicitly discussed in groundwater modeling applications to support decision making and avoid misinterpretation of modeling results.
ENVIRONMENTAL IMPACT ASSESSMENT REVIEW
(2021)
Article
Engineering, Civil
Nan Zhang, Xiaoguang Yang, Haifeng Guo, Hongzhao Dong, Wanjing Ma
Summary: Model-based traffic state estimation is used to reproduce traffic flow states at signalized intersections. A Bayesian learning framework is developed to learn unknown variables in the models from observed data. The proposed method shows good performance in estimating traffic flow at signalized intersections.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Biomedical
Mohammad R. Rezaei, Haseul Jeoung, Ayda Gharamani, Utpal Saha, Venkat Bhat, Milos R. Popovic, Ali Yousefi, Robert Chen, Milad Lankarany
Summary: In this study, a new model called HI-DGD is proposed to infer a cognitive state underlying decision-making based on neural activities and behavioral signals recorded from Parkinson's disease patients. The model demonstrates better performance in estimating the cognitive state of conflict and non-conflict trials compared to traditional methods, and identifies the specific neural features that contribute significantly to decision-making.
JOURNAL OF NEURAL ENGINEERING
(2023)
Article
Neurosciences
Michael W. Cole, Takuya Ito, Carrisa Cocuzza, Ruben Sanchez-Romero
Summary: Resting-state functional connectivity provides insights into brain network organization, but the functional importance of task-related changes remains unclear. Task-state functional connectivity can predict cognitive task activations better, driven by individual-specific functional connectivity patterns. These findings suggest task-related changes play a role in reshaping brain network organization and altering neural activity flow during task performance.
JOURNAL OF NEUROSCIENCE
(2021)
Article
Biology
Shuzhen Zuo, Lei Wang, Jung Han Shin, Yudian Cai, Boqiang Zhang, Sang Wan Lee, Kofi Appiah, Yong-di Zhou, Sze Chai Kwok
Article
Computer Science, Artificial Intelligence
Minryung R. Song, Sang Wan Lee
Review
Behavioral Sciences
John P. O'Doherty, Sang Wan Lee, Reza Tadayonnejad, Jeff Cockburn, Kyo Iigaya, Caroline J. Charpentier
Summary: The brain is proposed to act as a Mixture of Experts, with different expert systems proposing action strategies based on reliability. The anterior prefrontal cortex is suggested to play a specific role in this process, favoring simpler expert systems. Research indicates that this reliability-based control mechanism may be domain general.
NEUROSCIENCE AND BIOBEHAVIORAL REVIEWS
(2021)
Article
Biochemical Research Methods
Suyeon Heo, Yoondo Sung, Sang Wan Lee
Summary: The study revealed that subclinical depression affects model-based and model-free learning in the prefrontal-striatal circuitry, as well as disrupts the arbitration control between the two. Additionally, depression undermines the ability to exploit viable options, known as exploitation sensitivity. The findings suggest the potential for clinical applications, such as early diagnosis and behavioral therapy design, to address the impact of depression on decision-making processes.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Article
Cell Biology
Dongjae Kim, Jaeseung Jeong, Sang Wan Lee
Summary: The brain has been found to adaptively resolve the tradeoff between bias and variance during reinforcement learning, requiring baseline correction for prediction error to offset the adverse effects of irreducible error on value learning. Behavioral evidence of adaptive control has been shown in a Markov decision task with context changes, suggesting that the prediction error baseline signals context changes to improve adaptability. Multiplexed representations of prediction error baseline within specific brain regions have been identified, indicating their role in guiding model based and model-free reinforcement learning.
Article
Biochemical Research Methods
Jaejoong Kim, Sang Wan Lee, Seokho Yoon, Haeorm Park, Bumseok Jeong
Summary: This study investigates the neural computational mechanisms of controllability inference in a multi-agent setting. It reveals that information on others' action-outcome contingencies is integrated with one's own action-outcome contingencies to infer controllability, impacting motivated behavior. Additionally, a positive bias towards the self in multi-agent controllability inference is identified, affecting behavioral adaptation under volatile controllability.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yujin Cha, Sang Wan Lee
Summary: The study explores the possibility of accessing human uncertainty through deterministic neural networks and proposes a new model for human uncertainty inference. Experimental results demonstrate that the model can accurately predict both the uncertainty range and diagnoses given by humans, aiding in guiding human decision-making and facilitating more efficient and accurate learning.
THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Information Systems
Jongwoo Seo, Sang Wan Lee
Summary: The study found that neural network-based intuitive physics can successfully solve problems in non-inertial reference frames, which is an ability to understand and predict physical phenomena in advance. The research designed three experiments to represent different types of challenges and demonstrated that neural network methods can learn the underlying dynamics of objects from observations.
Proceedings Paper
Computer Science, Cybernetics
Dongjae Kim, Myeong Hyeon Kim, Sang Wan Lee
Summary: Recent studies have shown that learning strategies can be decoded from EEG data using a computational model, and the decoder may extract information applicable to various decision-making scenarios. The decoder contains a significant amount of mutual information between input, hidden, and output for both new and original training data, with informative features found in the model's deep layers for decoding decisions.
2021 9TH IEEE INTERNATIONAL WINTER CONFERENCE ON BRAIN-COMPUTER INTERFACE (BCI)
(2021)
Proceedings Paper
Acoustics
Mohamed Elgaar, Jungbae Park, Sang Wan Lee
2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Dongjae Kim, Jiseong Park, Jeongseok Hwang, Wan Hee Cho, Sang Wan Lee
2020 8TH INTERNATIONAL WINTER CONFERENCE ON BRAIN-COMPUTER INTERFACE (BCI)
(2020)
Article
Computer Science, Information Systems
Fengkai Ke, Seungjin Choi, Young Ho Kang, Keun-Ah Cheon, Sang Wan Lee
Proceedings Paper
Computer Science, Cybernetics
Dongjae Kim, Sang Wan Lee
2019 7TH INTERNATIONAL WINTER CONFERENCE ON BRAIN-COMPUTER INTERFACE (BCI)
(2019)
Proceedings Paper
Acoustics
Jungbae Park, Kijong Han, Yuneui Jeong, Sang Wan Lee
2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2019)
Proceedings Paper
Acoustics
Seokwon Jung, Jungbae Park, Sangwan Lee
2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2019)