Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

https://cancerbiologyprogram.med.wayne.edu/pdfs/hallmarks_cancer_article.pdf

Hallmarks of cancer: the next generation.

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. 
The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

In this letter, we report strong degradation of photoluminescence (PL) performance of GaN epilayers due to SiO2 layers that were deposited on GaN surfaces by electron-beam evaporation. Secondary ion mass spectrometry measurements show that the oxygen concentration of GaN with SiO2 layers is one order of magnitude more than that of as-grown GaN. This fact indicates that oxygen can very easily replace nitrogen in GaN. It was also found that rapid thermal processing can recover and improve the optical quality of GaN with SiO2 layer. As a reference, SixNy was found to have little effect on PL performance of GaN.

Strong influence of SiO2 thin film on properties of GaN epilayers

Biometric authentication involves various technologies to identify individuals by exploiting their unique, measurable physiological and behavioral characteristics. However, traditional biometric authentication systems (e.g., face recognition, iris, retina, voice, and fingerprint) are facing an increasing risk of being tricked by biometric tools such as anti-surveillance masks, contact lenses, vocoder, or fingerprint films. In this paper, we design a multimodal biometric authentication system named Deepkey, which uses both Electroencephalography (EEG) and gait signals to better protect against such risk. Deepkey consists of two key components: an Invalid ID Filter Model to block unauthorized subjects and an identification model based on attention-based Recurrent Neural Network (RNN) to identify a subject`s EEG IDs and gait IDs in parallel. The subject can only be granted access while all the components produce consistent evidence to match the user`s proclaimed identity. We implement Deepkey with a live deployment in our university and conduct extensive empirical experiments to study its technical feasibility in practice. DeepKey achieves the False Acceptance Rate (FAR) and the False Rejection Rate (FRR) of 0 and 1.0%, respectively. The preliminary results demonstrate that Deepkey is feasible, show consistent superior performance compared to a set of methods, and has the potential to be applied to the authentication deployment in real world settings.

DeepKey: An EEG and Gait Based Dual-Authentication System

Recently, two-dimensional (2D) materials have opened a new paradigm for fundamental physics explorations and device applications. Unlike gapless graphene, monolayer transition metal dichalcogenide (TMDC) has new optical functionalities for next generation ultra-compact electronic and opto-electronic devices. When TMDC crystals are thinned down to monolayers, they undergo an indirect to direct bandgap transition, making it an outstanding 2D semiconductor. Unique electron valley degree of freedom, strong light matter interactions and excitonic effects were observed. Enhancement of spontaneous emission has been reported on TMDC monolayers integrated with photonic crystal and distributed Bragg reflector microcavities. However, the coherent light emission from 2D monolayer TMDC has not been demonstrated, mainly due to that an atomic membrane has limited material gain volume and is lack of optical mode confinement. Here, we report the first realization of 2D excitonic laser by embedding monolayer tungsten disulfide (WS2) in a microdisk resonator. Using a whispering gallery mode (WGM) resonator with a high quality factor and optical confinement, we observed bright excitonic lasing in visible wavelength. The Si3N4/WS2/HSQ sandwich configuration provides a strong feedback and mode overlap with monolayer gain. This demonstration of 2D excitonic laser marks a major step towards 2D on-chip optoelectronics for high performance optical communication and computing applications.

Monolayer Excitonic Laser

/pdf/broadband-terahertz-silicon-membrane-metasurface-absorber-1ac9cwq3.pdf

Broadband Terahertz Silicon Membrane Metasurface Absorber

Matching of building polygons with different levels of detail is crucial in the maintenance and quality assessment of multi-representation databases. Two general problems need to be addressed in the matching process: (1) Which criteria are suitable? (2) How to effectively combine different criteria to make decisions? This paper mainly focuses on the second issue and views data matching as a supervised pattern classification. Several classifiers (i.e. decision trees, Naive Bayes and support vector machines) are evaluated for the matching task. Four criteria (i.e. position, size, shape and orientation) are used to extract information for these classifiers. Evidence shows that these classifiers outperformed the weighted average approach.

/pdf/pattern-classification-approaches-to-matching-building-38x2s7qvpy.pdf

Xiang Zhang

Papers

Strong influence of SiO2 thin film on properties of GaN epilayers

DeepKey: An EEG and Gait Based Dual-Authentication System

Monolayer Excitonic Laser

Broadband Terahertz Silicon Membrane Metasurface Absorber

Pattern classification approaches to matching building polygons at multiple scales