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

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

计量经济分析 = Econometric analysis

Sparse approximations in complex domain based on BM3D modeling

In this paper we present a novel wavelet-domain image upsampling algorithm based on iterative spatially adaptive filtering. A high-resolution image is reconstructed by alternating two procedures: spatially adaptive filtering and projection on the observation-constrained subspace. The Block-Matching and 3D filtering (BM3D) [3] technique is used to suppress ringing, and reconstruct missing wavelet detail coefficients. The BM3D algorithm exploits the local image statistics collected from similar blocks to extract local and non-local image features by 3D transform-domain shrinkage. It results in high-quality upsampled images, with sharp edges and practically no artifacts.

Image upsampling via spatially adaptive block-matching filtering

An algorithm for the blind separation of mutually independent and/or temporally correlated sources is presented in this letter. The algorithm is closely related to the maximum likelihood approach based on entropy rate minimization but uses a simpler contrast function that can be accurately and efficiently estimated using nearest-neighbor distances. The advantages of the new algorithm are highlighted using simulations and real electroencephalographic data.

Blind Source Separation by Entropy Rate Minimization

https://hal-univ-rennes1.archives-ouvertes.fr/hal-01240773/document

Efficiency of texture image enhancement by DCT-based filtering

A novel approach to image-denoising based on the shapeadaptive DCT (SA-DCT) is presented. The anisotropic LPAICI technique is used in order to deÞne the shape of the transforms support in a pointwise adaptive manner. It means that for each point in the image an adaptive estimation neighborhood is found. For each one of these neighborhoods a SADCT is performed. The thresholded SA-DCT coefÞcients are used to reconstruct a local estimate of the signal within the adaptive-shape region. Since regions corresponding to different points are in general overlapping (and thus generate an overcomplete representation of the signal), the local estimates are averaged together using adaptive weights that depend on the regions statistics. A Wiener Þltering procedure in SA-DCT domain is also proposed. Simulation experiments conÞrm the advanced quality of the Þnal estimate. Not only objective criteria scores are high, but also the visual appearence of the estimate is superior: edges are clean, and no unpleasant ringing artifacts are introduced by the Þtted transform.

Karen Egiazarian

Papers

Sparse approximations in complex domain based on BM3D modeling

Image upsampling via spatially adaptive block-matching filtering

Blind Source Separation by Entropy Rate Minimization

Efficiency of texture image enhancement by DCT-based filtering

Pointwise shape-adaptive dct as an overcomplete denoising tool