Search Results for author: Clarence W. de Silva

Found 9 papers, 5 papers with code

Backtracking Regression Forests for Accurate Camera Relocalization

1 code implementation22 Oct 2017 Lili Meng, Jianhui Chen, Frederick Tung, James J. Little, Julien Valentin, Clarence W. de Silva

Camera relocalization plays a vital role in many robotics and computer vision tasks, such as global localization, recovery from tracking failure, and loop closure detection.

Camera Relocalization Loop Closure Detection +2

Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features

1 code implementation22 Apr 2019 Kuangen Zhang, Ming Hao, Jing Wang, Clarence W. de Silva, Chenglong Fu

Learning on point cloud is eagerly in demand because the point cloud is a common type of geometric data and can aid robots to understand environments robustly.

Teach Biped Robots to Walk via Gait Principles and Reinforcement Learning with Adversarial Critics

1 code implementation22 Oct 2019 Kuangen Zhang, Zhimin Hou, Clarence W. de Silva, Haoyong Yu, Chenglong Fu

However, the local minima caused by unsuitable rewards and the overestimation of the cumulative reward impede the maximization of the cumulative reward.

Reinforcement Learning (RL)

Robust Fixed-Order Controller Design for Uncertain Systems with Generalized Common Lyapunov Strictly Positive Realness Characterization

no code implementations5 Jun 2020 Jun Ma, Haiyue Zhu, Xiaocong Li, Wenxin Wang, Clarence W. de Silva, Tong Heng Lee

It is also noteworthy that the proposed methodology additionally provides the necessary and sufficient conditions for this robust controller design with the consideration of a prescribed finite frequency range, and therefore significantly less conservatism is attained in the system performance.

Discriminative Feature Alignment: Improving Transferability of Unsupervised Domain Adaptation by Gaussian-guided Latent Alignment

1 code implementation23 Jun 2020 Jing Wang, Jiahong Chen, Jianzhe Lin, Leonid Sigal, Clarence W. de Silva

To solve this problem, we introduce a Gaussian-guided latent alignment approach to align the latent feature distributions of the two domains under the guidance of the prior distribution.

Data Augmentation Domain Generalization +3

How does the structure embedded in learning policy affect learning quadruped locomotion?

no code implementations29 Aug 2020 Kuangen Zhang, Jongwoo Lee, Zhimin Hou, Clarence W. de Silva, Chenglong Fu, Neville Hogan

This paper focuses on the latter because the structured policy is more intuitive and can inherit insights from previous model-based controllers.

Reinforcement Learning (RL)

Automated Multi-domain Engineering Design through Linear Graph and Genetic Programming

no code implementations25 Sep 2021 Eric McCormick, Haoxiang Lang, Clarence W. de Silva

This paper proposes a methodology of integrating the Linear Graph (LG) approach with Genetic Programming (GP) for generating an automated multi-domain engineering design approach by using the in-house developed LG MATLAB toolbox and the GP toolbox in MATLAB.

Preserving Domain Private Representation via Mutual Information Maximization

no code implementations9 Jan 2022 Jiahong Chen, Jing Wang, Weipeng Lin, Kuangen Zhang, Clarence W. de Silva

Recent advances in unsupervised domain adaptation have shown that mitigating the domain divergence by extracting the domain-invariant representation could significantly improve the generalization of a model to an unlabeled data domain.

Domain Generalization Unsupervised Domain Adaptation

Ensemble diverse hypotheses and knowledge distillation for unsupervised cross-subject adaptation

1 code implementation15 Apr 2022 Kuangen Zhang, Jiahong Chen, Jing Wang, Xinxing Chen, Yuquan Leng, Clarence W. de Silva, Chenglong Fu

EDH mitigates the divergence between labeled data of source subjects and unlabeled data of target subjects to accurately classify the locomotion modes of target subjects without labeling data.

Domain Adaptation Human Activity Recognition +1

Cannot find the paper you are looking for? You can Submit a new open access paper.