Machine learning for denoising by Quek, Kenneth Joo Hong

Get full text

Interactive learning on ECG by Zhu, Yu Ting

“…This report demonstrates the various interactive learning methods with the implementation of hardware components and software development. …”
Get full text

Multifidelity Methods for Design of Transition MetalComplexes by Janet, Jon Paul

“…Multiple sources of uncertainty that would limit the application of these methods to TM complexes are addressed. Surrogate models are trained to estimate system-specific DFT uncertainty by including data from DFT calculations with different fractions of exact exchange, and a novel uncertainty metric for data-driven discovery is proposed that quantifies the ability of ANNs to generalize to unseen data based on similarity in the learned latent space. …”
Get full text

From public health to AI safety: improving machine learning approaches by collecting, selecting, or reducing the need for high-quality data by Brauner, J

“…In the process, we develop novel machine learning (ML) methods to tackle various challenges. …”

The rise of deep learning in cyber security: Bibliometric analysis of deep learning and malware by Nur Khairani, Kamarudin, Ahmad Firdaus, Zainal Abidin, Mohd Zamri, Osman, Alanda, Alde, Erianda, Aldo, Shahreen, Kasim, Mohd Faizal, Ab Razak

“…Deep learning is a machine learning technology that allows computational models to learn via experience, mimicking human cognitive processes. …”
Get full text

MAS.714J / STS.445J Technologies for Creative Learning, Fall 2004 by Resnick, Mitchel

Subjects: “…learning…”
Get full text

Biometric contrastive learning for data-efficient deep learning from electrocardiographic images by Sangha, V, Khunte, A, Holste, G, Mortazavi, BJ, Wang, Z, Oikonomou, EK, Khera, R

“…We compared BCL with ImageNet initialization and general-purpose self-supervised contrastive learning for images (simCLR).</p> <p><strong>Results:&nbsp;</strong>While with 100% labeled training data, BCL performed similarly to other approaches for detecting AF/Gender/LVEF&thinsp;&lt;&thinsp;40% with an AUROC of 0.98/0.90/0.90 in the held-out test sets, it consistently outperformed other methods with smaller proportions of labeled data, reaching equivalent performance at 50% of data. …”

Visual recognition using deep learning (video captioning using deep learning) by Thong, Jing Lin

“…Thereafter, reinforcement learning techniques were used to further optimise the model. …”
Get full text

The devil is in the details: an evaluation of recent feature encoding methods by Chatfield, K, Lempitsky, V, Vedaldi, A, Zisserman, A

“…While several authors have reported very good results on the challenging PASCAL VOC classification data by means of these new techniques, differences in the feature computation and learning algorithms, missing details in the description of the methods, and different tuning of the various components, make it impossible to compare directly these methods and hard to reproduce the results reported. …”

Vision language representation learning by Yang, Xiaofeng

“…Despite its significance, learning effective vision language representation remains challenging due to several reasons. …”
Get full text

Optimization strategies for federated learning by Zhang, Tinghao

“…We achieve this through a deep reinforcement learning-based scheduling strategy and an optimized bandwidth allocation method, enabling FL to achieve target accuracy with reduced system costs. …”
Get full text

Fish classification and deep learning by Zhang, Dawei

Get full text

Transfer learning on UR robots by Yu, Xiwei

“…As neural networks and deep learning develop, researchers are continually exploring the capabilities and potential of using data-driven methods to control robots. …”
Get full text

Descriptor learning for efficient retrieval by Philbin, J, Isard, M, Sivic, J, Zisserman, A

“…Scalable, stochastic gradient methods are used for the optimization.</p> <br> <p>For the case of particular object retrieval, we demonstrate impressive gains in performance on a ground truth dataset: our learnt 32-D descriptor without spatial re-ranking outperforms a baseline method using 128-D SIFT descriptors with spatial re-ranking.…”

Robust partial-to-partial point cloud registration in a full range by Pan, Liang, Cai, Zhongang, Liu, Ziwei

“…Extensive experiments show that GMCNet outperforms previous state-of-the-art methods for PPR.…”
Get full text

Efficient rare event sampling with unsupervised normalizing flows by Asghar, Solomon, Pei, Qing-Xiang, Volpe, Giorgio, Ni, Ran

“…Classical computational methods to sample rare events remain prohibitively inefficient and are bottlenecks for enhanced samplers that require prior data. …”
Get full text

Reinforcement learning for robot assembly by Vuong Quoc Nghia

“…Finally, this thesis examines methods to narrow the reality gap - the fundamental problem in sim-to-real reinforcement learning. …”
Get full text

Image retrieval with deep learning by Tan, Joe Chin Yong

“…The query images are distorted with the 3 methods mentioned with different values of sigma, variance and quality. …”
Get full text

Online learning for search and classification by Nguyen, Thanh Tam

“…(i) Feature selection: we have investigated a number of newly supervised term weighting methods to improve the performance of text classification; (ii) Online classification: we have proposed several online learning algorithms that can be used for topic classification; (iii) Two-view online learning: we have proposed a two-view online learning algorithm, which can work on two-view datasets; (iv) Online learning-to-rank: for search engine, we have proposed an online learning-to-rank algorithm, which was to learn a scoring function to re-rank the search result.…”
Get full text

iTD3-CLN: learn to navigate in dynamic scene through Deep Reinforcement Learning by Jiang, Haoge, Esfahani, Mahdi Abolfazli, Wu, Keyu, Wan, Kong-wah, Heng, Kuan-kian, Wang, Han, Jiang, Xudong

“…In contrast to the conventional methods such as the DWA, our approach is found superior in the following ways: no need for prior knowledge of the environment and metric map, lower reliance on an accurate sensor, learning emergent behavior in dynamic scene that is intuitive, and more remarkably, able to transfer to the real robot without further fine-tuning. …”
Get full text