Counterfactual explanations for machine learning models on heterogeneous data by Wang, Yongjie

“…In real-life applications, it is difficult to compare changes in heterogeneous features with a scalar cost function. Moreover, existing methods do not support interactive exploration for users. …”
Get full text

Distributed formation for large swarm of agents by Reza Haghighi

“…In addition, two distributed shape formation control methodology is proposed to deal with bidirectional and unidirectional interactions among agents. The existing methods on shape formation control are limited to simple formation problems by single group of agents. …”
Get full text

One-class knowledge distillation for face presentation attack detection by Li, Zhi, Cai, Rizhao, Li, Haoliang, Lam, Kwok-Yan, Hu, Yongjian, Kot, Alex Chichung

“…Face presentation attack detection (PAD) has been extensively studied by research communities to enhance the security of face recognition systems. Although existing methods have achieved good performance on testing data with similar distribution as the training data, their performance degrades severely in application scenarios with data of unseen distributions. …”
Get full text

GPU-accelerated real-time motion planning for safe human-robot collaboration by Fujii, Shohei

“…Another challenge is that, to achieve true time-optimality in safe path tracking, it is crucial to have precise distances between obstacles and a robot at waypoints along an executing path. However, existing methods for computing distances between a robot and obstacles are either too slow for real-time applications, or inaccurate for achieving time-optimality. …”
Get full text

New Purification Strategies and Monomer Platforms for Ruthenium-Initiated Living Ring-Opening Metathesis Polymerization by Kilgallon, Landon J.

“…However, complete removal of ruthenium residues from the product is a difficult task that is compounded by the lack of understanding of initiator decomposition in ROMP. The existing methods for purification of ROMP polymers, which are typically solubility-based, are reviewed. …”
Get full text

Robust optimal operation of future power systems under uncertainties by Lahanda Purage Mohasha Isuru Sampath

“…As such, the scalability issues observed with existing methods are resolved. Further, a second-order cone relaxation is applied to nonlinear AC and DC NCs to improve the computational tractability and the globally optimal solvability. …”
Get full text

Toward a generic federated learning platform optimized for computer vision applications by Zhuang, Weiming

“…Extensive empirical studies demonstrate that our proposed approach is superior to existing methods in a wide range of settings. Moving forward from optimizing a single application, FL system could have multiple simultaneous training activities for multiple applications in real-world scenarios, such as autonomous vehicles. …”
Get full text

The relations between Van Der Waals and covalent forces in layered crystals by Chaturvedi, Apoorva

“…From the perspective of materials science, a novel method for rapid synthesis of thin few layer LTMD crystals has been proposed and developed in-house to complement other existing methods. The synthesis was achieved by plasma that was formed inside the inner sealed quartz tube heated by plasma surrounding the sealed ampoule. …”
Get full text

Real-time animation of vegetation : novel 2D approach and interactive authoring interface by Chen, Kan

“…It is not always easy or intuitive for users to generate their desired motions of short plants using existing methods. Thus, a novel approach to author the common motions of short plants which are bending and twisting has been proposed. …”
Get full text

Adversarial robustness of deep reinforcement learning by Qu, Xinghua

“…Our frame-correlation transfers dramatically speed up the attack generation compared to existing methods, often significantly reducing the required computation time; thus, shedding light on the real threat of real-time attacks in RL. …”
Get full text

Towards semantic, debiased and moment video retrieval by Satar, Burak

“…While many studies focus on improving pre-training or developing new backbones, existing methods may suffer from the learning and inference bias issue, as recent research suggests in other text-video-related tasks. …”
Get full text

Signal processing for cooperative cognitive radio networks by Rabiee, Ramtin

“…We aim to analyze some pre-existing methods such as improved energy detector (IED) or double-threshold comparison scheme in terms of detection performance and transmission throughput when a CSS-based detection scheme has been implemented among the secondary users. …”
Get full text

Online estimation of key battery states for management of vanadium redox flow battery by Wei, Zhongbao

“…The comparison with other existing methods also suggests its superiority in terms of accuracy, convergence speed, and computational cost. …”
Get full text

Advanced data analytics for future smart grid stability assessment by Ren, Chao

“…All the methodologies have been tested on benchmark power systems to verify their effectiveness, and comparative and comparative studies with existing methods in the literature are conducted where applicable.…”
Get full text

Fabrication and characterization of Algan/Gan high electron mobility transistors on silicon by Tham, Wai Hoe

“…Therefore, the current experimental studies suggest that the Ge doping approach is more suitable as a VTH tuning method. Although the fabrication cost of GaN-based devices can be reduced significantly through the incorporation with Si technology, there are several challenges which impede the incorporation of GaN-based devices with Si technology. …”
Get full text

Detection of road markings for advanced driver assistance by Suchitra Sathyanarayana

“…A study on how block size affects the compute efficiency was conducted to show that AHT reduces the total computation time by at least k2 times as compared to existing HT architectures for a k  k grid, translating to a speed up of at least 67 times against existing methods for an 8  8 grid. It was demonstrated that the block size in AHT and SLEM can be adjusted to achieve the required speedup and improve the quality of edge map, respectively. …”
Get full text

Kansei evaluation and clustering in emotional design by Ng, Kassyn.

“…The author modified an existing method, used generally in Kansei Engineering to cluster Kansei words. …”
Get full text

Deep reinforcement learning for optimal resource allocation by Ng, Steffi Si Yu

“…In addition, this project will practice parameter tuning methods on the algorithm to obtain a parameter set that can achieve the most optimal result.…”
Get full text

Placing objects in context via inpainting for out-of-distribution segmentation by De Jorge, P, Volpi, R, Dokania, PK, Torr, PHS, Rogez, G

“…In our experiments, we present different anomaly segmentation datasets based on POC-generated data and show that POC can improve the performance of recent state-of-the-art anomaly fine-tuning methods across several standardized benchmarks. …”

A size and position invariant event-based human posture recognition algorithm by Chen, Shoushun, Folowosele, Fopefolu, Kim, Dongsoo, Vogelstein, R. Jacob, Etienne-Cummings, Ralph, Culurciello, Eugenio

“…Compared to other existing method utilizing line segment Hausdorff distance, the proposed algorithm reduces the computation complexity by 36000 times, for our test images. …”
Get full text
Get full text