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

In this paper, we consider the problem of structure recovery in Markov Network over Gaussian variables, that is equivalent to finding the zero-pattern of the sparse inverse covariance matrix. Recently proposed l1-regularized optimization methods result into convex problems that can be solved optimally and efficiently. However, the accuracy such methods can be quite sensitive to the choice of regularization parameter, and optimal selection of this parameter remains an open problem. Herein, we adopt a Bayesian approach, treating the regularization parameter(s) as random variable(s) with some prior, and using MAP optimization to find both the inverse covariance matrix and the unknown regularization parameters. Our general formulation allows a vector of regularization parameters and is well-suited for learning structured graphs such as scale-free networks where the sparsity of nodes varies significantly. We present promising empirical results on both synthetic and real-life datasets, demonstrating that our approach achieves a better balance between the false-positive and falsenegative errors than commonly used approaches.

/pdf/sparse-markov-net-learning-with-priors-on-regularization-14e2rbnc75.pdf

Sparse Markov Net Learning with Priors on Regularization Parameters

Standard gradient descent algorithms applied to sequences of tasks are known to produce catastrophic forgetting in deep neural networks. When trained on a new task in a sequence, the model updates its parameters on the current task, forgetting past knowledge. This article explores scenarios where we scale the number of tasks in a ﬁnite environment. Those scenarios are composed of a long sequence of tasks with reoccurring data. We show that in such setting, stochastic gradient descent can learn, progress, and converge to a solution that according to existing literature needs a continual learning algorithm. In other words, we show that the model performs knowledge retention and accumulation without speciﬁc memorization mechanisms . We propose a new experimentation framework, SCoLe (Scaling Continual Learning), to study the knowledge retention and accumulation of algorithms in potentially inﬁnite sequences of tasks. To explore this setting, we performed a large number of experiments on sequences of 1,000 tasks to better understand this new family of settings. We also propose a slight modiﬁcations to the vanilla stochastic gradient descent to facilitate continual learning in this setting. The SCoLe framework represents a good simulation of practical training environments with reoccurring situations and allows the study of convergence behavior in long sequences. Our experiments show that previous results on short scenarios cannot always be extrapolated to longer scenarios.

Scaling the Number of Tasks in Continual Learning

A method, information processing system, and computer readable article of manufacture model data. A first dataset is received that includes a first set of physical world data. At least one data model associated with the first dataset is generated based on the receiving. A second dataset is received that includes a second set of physical world data. The second dataset is compared to the at least one data model. A probability that the second dataset is modeled by the at least one data model is determined. A determination is made that the probability is above a given threshold. A decision associated with the second dataset based on the at least one data model is generated in response to the probability being above the given threshold. The probability and the decision are stored in memory. The probability and the decision are provided to user via a user interface.

Recovering the structure of sparse markov networks from high-dimensional data

A unified framework is disclosed for representing and generating evaluation functions for a classification system. The disclosed unified framework provides evaluation functions having characteristics of both traditional or purity-based evaluation functions (class uniformity) and discrimination-based evaluation functions (discrimination power). The disclosed framework is based on a set of configurable parameters and is a function of the distance between examples. By varying the choice of parameters and the distance function, more emphasis is placed on either the class uniformity or the discrimination power of the induced example subsets. A user-configurable function is used to score each of the features based on the class uniformity and discrimination power measures and thereby select the feature having a highest score to partition the data (e.g., using a decision tree or rule-base). This process is recursively applied until all of the examples are partitioned.

Method and apparatus for representing and generating evaluation functions in a data classification system

The invariance principle from causality is at the heart of notable approaches such as invariant risk minimization (IRM) that seek to address out-of-distribution (OOD) generalization failures. Despite the promising theory, invariance principle-based approaches fail in common classification tasks, where invariant (causal) features capture all the information about the label. Are these failures due to the methods failing to capture the invariance? Or is the invariance principle itself insufficient? To answer these questions, we revisit the fundamental assumptions in linear regression tasks, where invariance-based approaches were shown to provably generalize OOD. In contrast to the linear regression tasks, we show that for linear classification tasks we need much stronger restrictions on the distribution shifts, or otherwise OOD generalization is impossible. Furthermore, even with appropriate restrictions on distribution shifts in place, we show that the invariance principle alone is insufficient. We prove that a form of the information bottleneck constraint along with invariance helps address key failures when invariant features capture all the information about the label and also retains the existing success when they do not. We propose an approach that incorporates both of these principles and demonstrate its effectiveness in several experiments.

Irina Rish

Papers

Sparse Markov Net Learning with Priors on Regularization Parameters

Scaling the Number of Tasks in Continual Learning

Recovering the structure of sparse markov networks from high-dimensional data

Method and apparatus for representing and generating evaluation functions in a data classification system

Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization