Xundong Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xundong Wu has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
Enrico Stefani
Division of Molecular Medicine, Department of Anesthesiology, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA
4 shared publications- Qing Wu
Hangzhou Dianzi University, Hangzhou, China
4 shared publications Ligia Toro
Division of Molecular Medicine, Department of Anesthesiology, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA
3 shared publications- Yong Wu
BH-428 CHS University of California 90095-7115 Los Angeles CA
3 shared publications - Chao Wang
Peking Univ, Coll Elect & Comp Engn, Shenzhen Grad Sch, Shenzhen 518055, Peoples R China
2 shared publications Yong Wu
2 shared publications
Latest Publications
- 9 September 2020
- Published by Springer Nature in Journal of Systems Science and Complexity
- Vol. 34 (1) , 68-82
Two-Stage Point Cloud Super Resolution with Local Interpolation and Readjustment via Outer-Product Neural Network
- 9 September 2020
- Published by Springer Nature in Journal of Systems Science and Complexity
- Vol. 34 (1) , 68-82
Two-Stage Point Cloud Super Resolution with Local Interpolation and Readjustment via Outer-Product Neural Network
- 1 August 2020
- Published by Elsevier in Neurocomputing
- Vol. 401, 113-122
SBNN: Slimming binarized neural network
- 1 August 2020
- Published by Elsevier in Neurocomputing
- Vol. 401, 113-122
SBNN: Slimming binarized neural network
- Open Access
Cross-Channel Intragroup Sparsity Neural Network
Modern deep neural networks rely on overparameterization to achieve state-of-the-art generalization. But overparameterized models are computationally expensive. Network pruning is often employed to obtain less demanding models for deployment. Fine-grained pruning removes individual weights in parameter tensors and can achieve a high model compression ratio with little accuracy degradation. However, it introduces irregularity into the computing dataflow and often does not yield improved model inference efficiency in practice. Coarse-grained model pruning, while realizing satisfactory inference speedup through removal of network weights in groups, e.g. an entire filter, often lead to significant accuracy degradation. This work introduces the cross-channel intragroup (CCI) sparsity structure, which can prevent the inference inefficiency of fine-grained pruning while maintaining outstanding model performance. We then present a novel training algorithm designed to perform well under the constraint imposed by the CCI-Sparsity. Through a series of comparative experiments we show that our proposed CCI-Sparsity structure and the corresponding pruning algorithm outperform prior art in inference efficiency by a substantial margin given suited hardware acceleration in the future.
Open AccessCross-Channel Intragroup Sparsity Neural Network
Modern deep neural networks rely on overparameterization to achieve state-of-the-art generalization. But overparameterized models are computationally expensive. Network pruning is often employed to obtain less demanding models for deployment. Fine-grained pruning removes individual weights in parameter tensors and can achieve a high model compression ratio with little accuracy degradation. However, it introduces irregularity into the computing dataflow and often does not yield improved model inference efficiency in practice. Coarse-grained model pruning, while realizing satisfactory inference speedup through removal of network weights in groups, e.g. an entire filter, often lead to significant accuracy degradation. This work introduces the cross-channel intragroup (CCI) sparsity structure, which can prevent the inference inefficiency of fine-grained pruning while maintaining outstanding model performance. We then present a novel training algorithm designed to perform well under the constraint imposed by the CCI-Sparsity. Through a series of comparative experiments we show that our proposed CCI-Sparsity structure and the corresponding pruning algorithm outperform prior art in inference efficiency by a substantial margin given suited hardware acceleration in the future.
- 1 June 2020
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 121, 103800
ECG signal classification with binarized convolutional neural network
- 1 June 2020
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 121, 103800
ECG signal classification with binarized convolutional neural network
- 9 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 17 (7) , 1278-1282
Improving the Accuracy of Binarized Neural Networks and Application on Remote Sensing Data
- 9 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 17 (7) , 1278-1282
Improving the Accuracy of Binarized Neural Networks and Application on Remote Sensing Data
- Open Access
- 3 May 2019
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 15 (5) , e1006892
How Dendrites Affect Online Recognition Memory
In order to record the stream of autobiographical information that defines our unique personal history, our brains must form durable memories from single brief exposures to the patterned stimuli that impinge on them continuously throughout life. However, little is known about the computational strategies or neural mechanisms that underlie the brain's ability to perform this type of "online" learning. Based on increasing evidence that dendrites act as both signaling and learning units in the brain, we developed an analytical model that relates online recognition memory capacity to roughly a dozen dendritic, network, pattern, and task-related parameters. We used the model to determine what dendrite size maximizes storage capacity under varying assumptions about pattern density and noise level. We show that over a several-fold range of both of these parameters, and over multiple orders-of-magnitude of memory size, capacity is maximized when dendrites contain a few hundred synapses—roughly the natural number found in memory-related areas of the brain. Thus, in comparison to entire neurons, dendrites increase storage capacity by providing a larger number of better-sized learning units. Our model provides the first normative theory that explains how dendrites increase the brain’s capacity for online learning; predicts which combinations of parameter settings we should expect to find in the brain under normal operating conditions; leads to novel interpretations of an array of existing experimental results; and provides a tool for understanding which changes associated with neurological disorders, aging, or stress are most likely to produce memory deficits—knowledge that could eventually help in the design of improved clinical treatments for memory loss. Humans can effortlessly recognize a pattern as familiar even after a single presentation and a long delay, and our capacity to do so even with complex stimuli such as images has been called "almost limitless". How is the information needed to support familiarity judgements stored so rapidly and held so reliably for such a long time? Most theoretical work aimed at understanding the brain's one-shot learning mechanisms has been based on drastically simplified neuron models which omit any representation of the most visually prominent features of neurons—their extensive dendritic arbors. Given recent evidence that individual dendritic branches generate local spikes, and function as separately thresholded learning/responding units inside neurons, we set out to capture mathematically how the numerous parameters needed to describe a dendrite-based neural learning system interact to determine the memory's storage capacity. Using the model, we show that having dendrite-sized learning units provides a large capacity boost compared to a memory based on simplified (dendriteless) neurons, attesting to the importance of dendrites for optimal memory function. Our mathematical model may also prove useful in future efforts to understand how disruptions to dendritic structure and function lead to reduced memory capacity in aging and disease.
- 3 May 2019
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 15 (5) , e1006892
Open AccessHow Dendrites Affect Online Recognition Memory
In order to record the stream of autobiographical information that defines our unique personal history, our brains must form durable memories from single brief exposures to the patterned stimuli that impinge on them continuously throughout life. However, little is known about the computational strategies or neural mechanisms that underlie the brain's ability to perform this type of "online" learning. Based on increasing evidence that dendrites act as both signaling and learning units in the brain, we developed an analytical model that relates online recognition memory capacity to roughly a dozen dendritic, network, pattern, and task-related parameters. We used the model to determine what dendrite size maximizes storage capacity under varying assumptions about pattern density and noise level. We show that over a several-fold range of both of these parameters, and over multiple orders-of-magnitude of memory size, capacity is maximized when dendrites contain a few hundred synapses—roughly the natural number found in memory-related areas of the brain. Thus, in comparison to entire neurons, dendrites increase storage capacity by providing a larger number of better-sized learning units. Our model provides the first normative theory that explains how dendrites increase the brain’s capacity for online learning; predicts which combinations of parameter settings we should expect to find in the brain under normal operating conditions; leads to novel interpretations of an array of existing experimental results; and provides a tool for understanding which changes associated with neurological disorders, aging, or stress are most likely to produce memory deficits—knowledge that could eventually help in the design of improved clinical treatments for memory loss. Humans can effortlessly recognize a pattern as familiar even after a single presentation and a long delay, and our capacity to do so even with complex stimuli such as images has been called "almost limitless". How is the information needed to support familiarity judgements stored so rapidly and held so reliably for such a long time? Most theoretical work aimed at understanding the brain's one-shot learning mechanisms has been based on drastically simplified neuron models which omit any representation of the most visually prominent features of neurons—their extensive dendritic arbors. Given recent evidence that individual dendritic branches generate local spikes, and function as separately thresholded learning/responding units inside neurons, we set out to capture mathematically how the numerous parameters needed to describe a dendrite-based neural learning system interact to determine the memory's storage capacity. Using the model, we show that having dendrite-sized learning units provides a large capacity boost compared to a memory based on simplified (dendriteless) neurons, attesting to the importance of dendrites for optimal memory function. Our mathematical model may also prove useful in future efforts to understand how disruptions to dendritic structure and function lead to reduced memory capacity in aging and disease.
- Open Access
Binarized Neural Networks on the ImageNet Classification Task
We trained Binarized Neural Networks (BNNs) on the high resolution ImageNet ILSVRC-2102 dataset classification task and achieved a good performance. With a moderate size network of 13 layers, we obtained top-5 classification accuracy rate of 84.1 % on validation set through network distillation, much better than previous published results of 73.2% on XNOR network and 69.1% on binarized GoogleNET. We expect networks of better performance can be obtained by following our current strategies. We provide a detailed discussion and preliminary analysis on strategies used in the network training.
Open AccessBinarized Neural Networks on the ImageNet Classification Task
We trained Binarized Neural Networks (BNNs) on the high resolution ImageNet ILSVRC-2102 dataset classification task and achieved a good performance. With a moderate size network of 13 layers, we obtained top-5 classification accuracy rate of 84.1 % on validation set through network distillation, much better than previous published results of 73.2% on XNOR network and 69.1% on binarized GoogleNET. We expect networks of better performance can be obtained by following our current strategies. We provide a detailed discussion and preliminary analysis on strategies used in the network training.
- Open Access
- 1 October 2015
- Published by Springer Nature in Scientific Reports
- Vol. 5 (1) , 14766
Resonant Scanning with Large Field of View Reduces Photobleaching and Enhances Fluorescence Yield in STED Microscopy
Photobleaching is a major limitation of superresolution Stimulated Depletion Emission (STED) microscopy. Fast scanning has long been considered an effective means to reduce photobleaching in fluorescence microscopy, but a careful quantitative study of this issue is missing. In this paper, we show that the photobleaching rate in STED microscopy can be slowed down and the fluorescence yield be enhanced by scanning with high speed, enabled by using large field of view in a custom-built resonant-scanning STED microscope. The effect of scanning speed on photobleaching and fluorescence yield is more remarkable at higher levels of depletion laser irradiance and virtually disappears in conventional confocal microscopy. With ≥6 GW∙cm−2 depletion irradiance, we were able to extend the fluorophore survival time of Atto 647N and Abberior STAR 635P by ~80% with 8-fold wider field of view. We confirm that STED Photobleaching is primarily caused by the depletion light acting upon the excited fluorophores. Experimental data agree with a theoretical model. Our results encourage further increasing the linear scanning speed for photobleaching reduction in STED microscopy.
- 1 October 2015
- Published by Springer Nature in Scientific Reports
- Vol. 5 (1) , 14766
Open AccessResonant Scanning with Large Field of View Reduces Photobleaching and Enhances Fluorescence Yield in STED Microscopy
Photobleaching is a major limitation of superresolution Stimulated Depletion Emission (STED) microscopy. Fast scanning has long been considered an effective means to reduce photobleaching in fluorescence microscopy, but a careful quantitative study of this issue is missing. In this paper, we show that the photobleaching rate in STED microscopy can be slowed down and the fluorescence yield be enhanced by scanning with high speed, enabled by using large field of view in a custom-built resonant-scanning STED microscope. The effect of scanning speed on photobleaching and fluorescence yield is more remarkable at higher levels of depletion laser irradiance and virtually disappears in conventional confocal microscopy. With ≥6 GW∙cm−2 depletion irradiance, we were able to extend the fluorophore survival time of Atto 647N and Abberior STAR 635P by ~80% with 8-fold wider field of view. We confirm that STED Photobleaching is primarily caused by the depletion light acting upon the excited fluorophores. Experimental data agree with a theoretical model. Our results encourage further increasing the linear scanning speed for photobleaching reduction in STED microscopy.