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

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

Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.

/pdf/a-survey-on-image-data-augmentation-for-deep-learning-4ru1bj2hbb.pdf

A survey on Image Data Augmentation for Deep Learning

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

Deep artificial neural networks require a large corpus of training data in order to effectively learn, where collection of such training data is often expensive and laborious.Data augmentationovercomes this issue by artificially inflating the training set with label preserving transformations. Recently there has been extensive use of generic data augmentation to improveConvolutional Neural Network(CNN) task performance. This study benchmarks various popular data augmentation schemes to allow researchers to make informed decisions as to which training methods are most appropriate for their data sets. Various geometric and photometric schemes are evaluated on a coarse-grained data set using a relatively simple CNN. Experimental results, run using 4-fold cross-validation and reported in terms of Top-1 and Top-5 accuracy, indicate that croppingin geometric augmentationsignificantly increases CNN task performance.

Improving Deep Learning with Generic Data Augmentation

Deep artificial neural networks require a large corpus of training data in order to effectively learn, where collection of such training data is often expensive and laborious. Data augmentation overcomes this issue by artificially inflating the training set with label preserving transformations. Recently there has been extensive use of generic data augmentation to improve Convolutional Neural Network (CNN) task performance. This study benchmarks various popular data augmentation schemes to allow researchers to make informed decisions as to which training methods are most appropriate for their data sets. Various geometric and photometric schemes are evaluated on a coarse-grained data set using a relatively simple CNN. Experimental results, run using 4-fold cross-validation and reported in terms of Top-1 and Top-5 accuracy, indicate that cropping in geometric augmentation significantly increases CNN task performance.

Improving Deep Learning using Generic Data Augmentation

This article comparatively tests three cooperative co-evolution methods for automated controller design in simulated robot teams. Collective NeuroEvolution (CONE) co-evolves multiple robot controllers using emergent behavioral specialization in order to increase collective behavior task performance. CONE is comparatively evaluated with two related controller design methods in a collective construction task. The task requires robots to gather building blocks and assemble the blocks in specific sequences in order to build structures. Results indicate that for the team sizes tested, CONE yields a higher collective behavior task performance (comparative to related methods) as a consequence of its capability to evolve specialized behaviors.

https://sacral.c.u-tokyo.ac.jp/pdf/SaEC-NitschkeSchutEiben-2011.pdf

Evolving behavioral specialization in robot teams to solve a collective construction task

This review presents a review of prevalent results within research pertaining to emergent cooperation in biologically inspired artificial social systems. Results reviewed maintain particular reference to biologically inspired design principles, given that current mathematical and empirical tools have provided only a partial insight into elucidating mechanisms responsible for emergent cooperation, and then only in systems of an abstract nature. This review aims to provide an overview of important and disparate research contributions that investigate utilization of biologically inspired concepts such as emergence, evolution, and self-organization as a means of attaining cooperation in artificial social systems. An introduction and overview of emergent cooperation in artificial life is presented, followed by a survey of emergent cooperation in swarm-based systems, the pursuit-evasion domain, and RoboCup soccer. The final section draws conclusions regarding future directions of emergent cooperation as a problem-solving methodology that is potentially applicable in a wide range of problem domains. Within each of these sections and their respective themes of research, the mechanisms deemed to be responsible for emergent cooperation are elucidated and their key limitations highlighted. The review concludes that current studies in emergent cooperative behavior are limited by a lack of situated and embodied approaches, and by the research infancy of current biologically inspired design approaches. Despite these limiting factors, emergent cooperation maintains considerable future potential in a wide variety of application domains where systems composed of many interacting components must cooperatively perform unanticipated global tasks.

/pdf/emergence-of-cooperation-state-of-the-art-59fz4jsvov.pdf

Emergence of Cooperation: State of the Art

Specialization is observable in many complex adaptive systems and is thought by many to be a fundamental mechanism for achieving optimal efficiency within organizations operating within complex adaptive systems. This chapter presents a survey and critique of collective behavior systems designed using biologically inspired principles, where specialization that emerges as a result of system dynamics and is used problem solver or means to increase task performance. The chapter presents an argument for developing design methodologies and principles that facilitate emergent specialization in collective behavior systems. Open problems of current research and future research directions are highlighted for the purpose of encouraging the development of such emergent specialization design methodologies.

/pdf/emergent-specialization-in-biologically-inspired-collective-3ti7lm1fla.pdf

Geoff Nitschke

Papers

Improving Deep Learning with Generic Data Augmentation

Improving Deep Learning using Generic Data Augmentation

Evolving behavioral specialization in robot teams to solve a collective construction task

Emergence of Cooperation: State of the Art

Emergent Specialization in Biologically Inspired Collective Behavior Systems