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

/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

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

Statistical Analysis with Missing Data

An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

Various embodiments select features from a feature space. In one embodiment, a set of training samples and a set of test samples are received. A first centered Gram matrix of a given dimension is determined for each of a set of feature vectors that include at least one of the set of training samples and at least one of the set of test samples. A second centered Gram matrix of the given dimension is determined for a target value vector that includes target values from the set of training samples. A set of columns and rows associated with the at least one of the test samples in the second centered Gram matrix is set to 0. A subset of features is selected from a set of features based on the first and second centered Gram matrices.

Transductive lasso for high-dimensional data regression problems

A method, computer readable storage medium, computer program product and a service. A profile of a computer user is obtained that contains meta tags descriptive of the participants of a first social networking website. A second social networking website having meta tags is selected. A profile from the second social networking website is selected. The meta tags of the first and second social networking websites are compared to determine if there is a match of at least one meta tag. Then, a search is made for related websites having at least one meta tag that matches the at least one meta tag. A list of the related websites is then presented to the computer user.

Presentation of websites to a computer user

We consider the problem of diagnosing performance problems in distributed system and networks given end-to-end performance measurements provided by test transactions, or probes. Common techniques for problem diagnosis such as, for example, codebook and network tomography usually assume a known dependency (e.g., routing) matrix that describes how each probe depends on the systems components. However, collecting full information about routing and/or probe dependencies on all systems components can be very costly, if not impossible, in large-scale, dynamic networks and distributed systems. We propose an approach to problem diagnosis and dependency discovery from end-to-end performance measurements in cases when the dependency/routing information is unknown or partially known. Our method is based on Blind Source Separation (BSS) approach that aims at reconstructing unobserved input signals and the mixing-weights matrix from the observed mixtures of signals. Particularly, we apply sparse non-negative matrix factorization techniques that appear particularly fitted to the problem of recovering network bottlenecks and dependency (routing) matrix, and show promising experimental results on several realistic network topologies.

/pdf/blind-source-separation-approach-to-performance-diagnosis-2bvazaczgu.pdf

Blind source separation approach to performance diagnosis and dependency discovery

Neural networks are known to be vulnerable to adversarial attacks -- slight but carefully constructed perturbations of the inputs which can drastically impair the network's performance. Many defense methods have been proposed for improving robustness of deep networks by training them on adversarially perturbed inputs. However, these models often remain vulnerable to new types of attacks not seen during training, and even to slightly stronger versions of previously seen attacks. In this work, we propose a novel approach to adversarial robustness, which builds upon the insights from the domain adaptation field. Our method, called Adversarial Feature Desensitization (AFD), aims at learning features that are invariant towards adversarial perturbations of the inputs. This is achieved through a game where we learn features that are both predictive and robust (insensitive to adversarial attacks), i.e. cannot be used to discriminate between natural and adversarial data. Empirical results on several benchmarks demonstrate the effectiveness of the proposed approach against a wide range of attack types and attack strengths.

Adversarial Feature Desensitization

Federated learning is an emerging paradigm that permits a large number of clients with heterogeneous data to coordinate learning of a unified global model without the need to share data amongst each other. Standard federated learning algorithms involve averaging of model parameters or gradient updates to approximate the global model at the server. However, in heterogeneous settings averaging can result in information loss and lead to poor generalization due to the bias induced by dominant clients. We hypothesize that to generalize better across non-i.i.d datasets as in FL settings, the algorithms should focus on learning the invariant mechanism that is constant while ignoring spurious mechanisms that differ across clients. Inspired from recent work in the Out-of-Distribution literature, we propose a gradient masked averaging approach for federated learning as an alternative to the standard averaging of client updates. This client update aggregation technique can be adapted as a drop-in replacement in most existing federated algorithms. We perform extensive experiments with gradient masked approach on multiple FL algorithms with in-distribution, real-world, and out-of-distribution (as the worst case scenario) test dataset and show that it provides consistent improvements, particularly in the case of heterogeneous clients.

Irina Rish

Papers

Transductive lasso for high-dimensional data regression problems

Presentation of websites to a computer user

Blind source separation approach to performance diagnosis and dependency discovery

Adversarial Feature Desensitization

Gradient Masked Averaging for Federated Learning