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

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

There are a huge number of features which are said to improve Convolutional Neural Network (CNN) accuracy. Practical testing of combinations of such features on large datasets, and theoretical justification of the result, is required. Some features operate on certain models exclusively and for certain problems exclusively, or only for small-scale datasets; while some features, such as batch-normalization and residual-connections, are applicable to the majority of models, tasks, and datasets. We assume that such universal features include Weighted-Residual-Connections (WRC), Cross-Stage-Partial-connections (CSP), Cross mini-Batch Normalization (CmBN), Self-adversarial-training (SAT) and Mish-activation. We use new features: WRC, CSP, CmBN, SAT, Mish activation, Mosaic data augmentation, CmBN, DropBlock regularization, and CIoU loss, and combine some of them to achieve state-of-the-art results: 43.5% AP (65.7% AP50) for the MS COCO dataset at a realtime speed of ~65 FPS on Tesla V100. Source code is at this https URL

YOLOv4: Optimal Speed and Accuracy of Object Detection

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

In this paper, a novel robust fault-tolerant control scheme based on event-triggered communication mechanism for a variable-speed wind energy conversion system (WECS) with sensor and actuator failures is proposed. The nonlinear WECS with event-triggered mechanism is modeled based on the Takagi–Sugeno (T–S) fuzzy model. By Lyapunov stability theory, the parameter expression of the proposed robust fault-tolerant controller with event-triggered mechanisms is proposed based on a feasible solution of linear matrix inequalities. Compared with the existing WECS fault-tolerant control methods, the proposed scheme significantly reduces the pressure of network packet transmission and improves the robustness and reliability of the WECS. Considering a doubly-fed variable speed constant frequency wind turbine, the event- triggered mechanism based fault-tolerant control for WECS is analyzed considering system model uncertainty. Numerical simulation results demonstrate that the proposed scheme is feasible and effective. 

Event-triggered mechanism based robust fault-tolerant control for networked wind energy conversion system

Discovery of novel benzofuro[3,2-b]quinoline derivatives as dual CDK2/Topo I inhibitors.

Separation and single-cell analysis for free gastric cancer cells in ascites and peritoneal lavages based on microfluidic chips

We describe the problem of classifying hundreds of millions of Wi-Fi hotspots using only connection and user count characteristics. We use a combination of deep learning and frequency analysis. Specifically, Convolution Neural Networks (CNN) capture the spatio-temporal relationship between adjacent connection/user counts across a 24 hour×7 day matrix, while FFT (Fast Fourier Transforms) extract user and connection frequencies. Our production system has been deployed to classify 239 million hotspots in 12 hours on a SPARK 2.0 cluster, achieving close to 80% F1-score for binary classification.

Large-Scale Wi-Fi Hotspot Classification via Deep Learning

CDK2/9 are members of the CDKs family, which play key roles in the occurrence and development of many cancers by regulating cell cycle and transcriptional prolongation, respectively. To further optimize and discuss the structure-activity relationships (SARs), a series of tacrine-based compounds were designed and synthesized from the compound ZLWT-37, which was studied by our group previously but no detailed SARs study was conducted on CDK2/9. Among this series, compounds ZLMT-12 (35) exhibited the most potent antiproliferative activity (GI50 = 0.006 μM for HCT116) and superior CDK2/9 inhibitory properties (CDK2: IC50 = 0.011 μM, CDK9: IC50 = 0.002 μM). Meanwhile, ZLMT-12 showed a weak inhibitory effect on acetylcholinesterase (AChE, IC50 = 19.023 μM) and butyrylcholinesterase (BuChE, IC50 = 2.768 μM). In addition, ZLMT-12 can suppress colony formation and migration in HCT116 cells, as well as induce the apoptosis and arrest the cell cycle in the S phase and G2/M phase. In vivo investigations revealed that ZLMT-12 inhibits tumor growth in the HCT116 xenograft tumor model at a low dose of 10 mg/kg without causing hepatotoxicity. The acute toxicity test showed low toxicity with a median lethal dosage (LD50) of 104.417 mg/kg. These findings showed that ZLMT-12 might be used as a drug candidate by targeting CDK2/9. 

Chang Xu

Papers

Event-triggered mechanism based robust fault-tolerant control for networked wind energy conversion system

Discovery of novel benzofuro[3,2-b]quinoline derivatives as dual CDK2/Topo I inhibitors.

Separation and single-cell analysis for free gastric cancer cells in ascites and peritoneal lavages based on microfluidic chips

Large-Scale Wi-Fi Hotspot Classification via Deep Learning

Development and structure-activity relationship of tacrine derivatives as highly potent CDK2/9 inhibitors for the treatment of cancer.