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

There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

A major assumption in many machine learning and data mining algorithms is that the training and future data must be in the same feature space and have the same distribution. However, in many real-world applications, this assumption may not hold. For example, we sometimes have a classification task in one domain of interest, but we only have sufficient training data in another domain of interest, where the latter data may be in a different feature space or follow a different data distribution. In such cases, knowledge transfer, if done successfully, would greatly improve the performance of learning by avoiding much expensive data-labeling efforts. In recent years, transfer learning has emerged as a new learning framework to address this problem. This survey focuses on categorizing and reviewing the current progress on transfer learning for classification, regression, and clustering problems. In this survey, we discuss the relationship between transfer learning and other related machine learning techniques such as domain adaptation, multitask learning and sample selection bias, as well as covariate shift. We also explore some potential future issues in transfer learning research.

/pdf/a-survey-on-transfer-learning-1hjmu3otql.pdf

A Survey on Transfer Learning

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

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

We describe a novel neural network architecture, which can recognize human faces with any view in a certain viewing angle range (from left 30 degrees to right 30 degrees out of plane rotation). View-specific eigenface analysis is used as the front-end of the system to extract features, and the neural network ensemble is used for recognition. Experimental results show that the recognition accuracy of our network ensemble is higher than conventional methods such as using a single neural network to recognize faces of a specific view.

/pdf/pose-invariant-face-recognition-1v297b18fq.pdf

Pose invariant face recognition

The traditional setting of supervised learning requires a large amount of labeled training examples in order to achieve good generalization. However, in many practical applications, unlabeled training examples are readily available, but labeled ones are fairly expensive to obtain. Therefore, semisupervised learning has attracted much attention. Previous research on semisupervised learning mainly focuses on semisupervised classification. Although regression is almost as important as classification, semisupervised regression is largely understudied. In particular, although cotraining is a main paradigm in semisupervised learning, few works has been devoted to cotraining-style semisupervised regression algorithms. In this paper, a cotraining-style semisupervised regression algorithm, that is, COREG, is proposed. This algorithm uses two regressors, each labels the unlabeled data for the other regressor, where the confidence in labeling an unlabeled example is estimated through the amount of reduction in mean squared error over the labeled neighborhood of that example. Analysis and experiments show that COREG can effectively exploit unlabeled data to improve regression estimates.

Semisupervised Regression with Cotraining-Style Algorithms

In semi-supervised learning, a number of labeled examples are usually required for training an initial weakly useful predictor which is in turn used for exploiting the unlabeled examples. However, in many real-world applications there may exist very few labeled training examples, which makes the weakly useful predictor difficult to generate, and therefore these semisupervised learning methods cannot be applied. This paper proposes a method working under a two-view setting. By taking advantages of the correlations between the views using canonical component analysis, the proposed method can perform semi-supervised learning with only one labeled training example. Experiments and an application to content-based image retrieval validate the effectiveness of the proposed method.

/pdf/semi-supervised-learning-with-very-few-labeled-training-4tqhwyabpi.pdf

Semi-supervised learning with very few labeled training examples

Spectral clustering (SC) is a large family of grouping methods that partition data using eigenvectors of an affinity matrix derived from the data. Though SC methods have been successfully applied to a large number of challenging clustering scenarios, it is noteworthy that they will fail when there are significant intersections among different clusters. In this paper, based on the analysis that SC methods are able to work well when the affinity values of the points belonging to different clusters are relatively low, we propose a new method, called spectral multi-manifold clustering (SMMC), which is able to handle intersections. In our model, the data are assumed to lie on or close to multiple smooth low-dimensional manifolds, where some data manifolds are separated but some are intersecting. Then, local geometric information of the sampled data is incorporated to construct a suitable affinity matrix. Finally, spectral method is applied to this affinity matrix to group the data. Extensive experiments on synthetic as well as real datasets demonstrate the promising performance of SMMC.

https://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/tnn11smmc.pdf

Spectral Clustering on Multiple Manifolds

Neural network ensemble is a learning paradigm where several neural networks are jointly used to solve a problem. In this paper, the relationship between the generalization ability of the neural network ensemble and the correlation of the individual neural networks is analyzed, which reveals that ensembling a selective subset of individual networks is superior to ensembling all the individual networks in some cases. Therefore an approach named GASEN is proposed, which trains several individual neural networks and then employs genetic algorithm to select an optimum subset of individual networks to constitute an ensemble. Experimental results show that, comparing with a popular ensemble approach, i.e. averaging all, and a theoretically optimum selective ensemble approach, i.e. enumerating, GASEN has preferable performance in generating ensembles with strong generalization ability in relatively small computational cost.

https://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/ijcai01.pdf

Zhi-Hua Zhou

Papers

Pose invariant face recognition

Semisupervised Regression with Cotraining-Style Algorithms

Semi-supervised learning with very few labeled training examples

Spectral Clustering on Multiple Manifolds

Genetic algorithm based selective neural network ensemble