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

This article presents a survey of prevalent research results pertaining to emergent cooperation in RoboCup soccer. Results reviewed maintain particular reference to research that uses biologically inspired design principles and concepts, such as emergence and evolution, as a means of attaining cooperative behavior. The core of this article argues that even though emergent cooperative behavior derived within RoboCup (and the larger field of multi-robot systems) is still in its infancy, it holds considerable future potential, as a problem solver in domains where systems comprising many interacting components must cooperatively solve a global task.

Emergent cooperation in robocup : A review

Robustness and evolvability have traditionally been seen as conflicting properties of evolutionary systems, due to the fact that selection requires heritable variation on which to operate. Various recent studies have demonstrated that organisms evolving in environments fluctuating non-randomly become better at adapting to these fluctuations, that is, increase their evolvability. It has been suggested that this is due to the emergence of biases in the mutational neighborhoods of genotypes. This paper examines a potential consequence of these observations, that a large bias in certain areas of genotype space will lead to increased robustness in corresponding phenotypes. The evolution of boolean networks, which bear similarity to models of gene regulatory networks, is simulated in environments which fluctuate between task targets. It was found that an increase in evolvability is concomitant with the emergence of highly robust genotypes, where evolvability was defined as the populations adaptability. Analysis of the genotype space elucidated that evolution finds regions containing robust genotypes coding for one of the target phenotypes, where these regions overlap or are situated in close proximity. Results indicate that genotype space topology impacts the relationship between robustness and evolvability, where the separation of robust regions coding for the various targets was detrimental to evolvability.

The Relationship Between Evolvability and Robustness in the Evolution of Boolean Networks

This paper reports upon two adaptive approaches for deriving words in an artificial language simulation. The efficacy of a Particle Swarm Optimization (PSO) method versus an Artificial Evolution (AE) method was examined for the purpose of adapting communication between agents. The objective of the study was for agents to derive a common (shared) lexicon for talking about food resources in the simulation environment. In the simulation, communication was essential for agent survival and as such facilitated lexicon adaptation. Results indicated that PSO was effective at adapting agents to quickly converge to a common lexicon, where, on average, one word for each food type was derived. AE required more method iterations to converge to a common lexicon that contained, on average, multiple words for each food type. However, there was greater word diversity in the lexicon converged upon by AE evolved agents, compared to that converged upon by PSO adapted agents.

Evolutionary algorithms and Particle Swarm Optimization for artificial language evolution

In this research, a neuro-evolution method called collective neuro-evolution (CONE), is introduced for the design of neural controllers (agents) operating in collective behavior task domains. The efficacy of the CONE method for facilitating emergent behavioral specialization for the benefit of increasing task performance is tested in a pursuit-evasion and collective gathering task. For a comparative study, a conventional neuro-evolution method was applied to the same tasks. In both tasks, the CONE method derived behavioral specialization in groups of agents resulting in higher task performances, where as the conventional neuro-evolution method was unable to derive specialization resulting in comparatively lower task performances

/pdf/evolutionary-design-of-specialization-4o4jh50oj2.pdf

Evolutionary Design of Specialization

This research concerns the comparison of three different artificial evolution approaches to the design of cooperative behavior in a group of simulated mobile robots. The first and second approaches, termed: single pool and plasticity, are characterized by robots that share a single genotype, though the plasticity approach includes a learning mechanism. The third approach, termed: multiple pools, is characterized by robots that use different genotypes. The application domain implements a pursuit-evasion game in which teams of robots of various sizes, termed: predators, collectively work to capture either one or two others, termed: prey. These artificial evolution approaches are also compared with a static rule based cooperative pursuit strategy specified a priori. Results indicate that the multiple pools approach is superior comparative to the other approaches in terms of measures defined for prey-capture strategy performance. That is, this approach facilitated specialization of behavioral roles allowing it to be effective for all predator team sizes tested.

/pdf/emergence-of-cooperation-in-a-pursuit-evasion-game-307rhu1upm.pdf

Geoff Nitschke

Papers

Emergent cooperation in robocup : A review

The Relationship Between Evolvability and Robustness in the Evolution of Boolean Networks

Evolutionary algorithms and Particle Swarm Optimization for artificial language evolution

Evolutionary Design of Specialization

Emergence of cooperation in a pursuit-evasion game