Intrinsically Motivated Learning In Natural And Artificial Systems

It has become clear to researchers in robotics and adaptive behaviour that current approaches are yielding systems with limited autonomy and capacity for self-improvement. To learn autonomously and in a cumulative fashion is one of the hallmarks of intelligence, and we know that higher mammals engage in exploratory activities that are not directed to pursue goals of immediate relevance for survival and reproduction but are instead driven by intrinsic motivations such as curiosity, interest in novel stimuli or surprising events, and interest in learning new behaviours. The adaptive value of such intrinsically motivated activities lies in the fact that they allow the cumulative acquisition of knowledge and skills that can be used later to accomplish fitness-enhancing goals. Intrinsic motivations continue during adulthood, and in humans they underlie lifelong learning, artistic creativity, and scientific discovery, while they are also the basis for processes that strongly affect human well-being, such as the sense of competence, self-determination, and self-esteem. This book has two aims: to present the state of the art in research on intrinsically motivated learning, and to identify the related scientific and technological open challenges and most promising research directions. The book introduces the concept of intrinsic motivation in artificial systems, reviews the relevant literature, offers insights from the neural and behavioural sciences, and presents novel tools for research. The book is organized into six parts: the chapters in Part I give general overviews on the concept of intrinsic motivations, their function, and possible mechanisms for implementing them; Parts II, III, and IV focus on three classes of intrinsic motivation mechanisms, those based on predictors, on novelty, and on competence; Part V discusses mechanisms that are complementary to intrinsic motivations; and Part VI introduces tools and experimental frameworks for investigating intrinsic motivations. The contributing authors are among the pioneers carrying out fundamental work on this topic, drawn from related disciplines such as artificial intelligence, robotics, artificial life, evolution, machine learning, developmental psychology, cognitive science, and neuroscience. The book will be of value to graduate students and academic researchers in these domains, and to engineers engaged with the design of autonomous, adaptive robots. The contributing authors are among the pioneers carrying out fundamental work on this topic, drawn from related disciplines such as artificial intelligence, robotics, artificial life, evolution, machine learning, developmental psychology, cognitive science, and neuroscience. The book will be of value to graduate students and academic researchers in these domains, and to engineers engaged with the design of autonomous, adaptive robots.

The aim of this Research Topic for Frontiers in Psychology under the section of Cognitive Science and Frontiers in Neurorobotics is to present state-of-the-art research, whether theoretical, empirical, or computational investigations, on open-ended development driven by intrinsic motivations. The topic will address questions such as: How do motivations drive learning? How are complex skills built up from a foundation of simpler competencies? What are the neural and computational bases for intrinsically motivated learning? What is the contribution of intrinsic motivations to wider cognition? Autonomous development and lifelong open-ended learning are hallmarks of intelligence. Higher mammals, and especially humans, engage in activities that do not appear to directly serve the goals of survival, reproduction, or material advantage. Rather, a large part of their activity is intrinsically motivated - behavior driven by curiosity, play, interest in novel stimuli and surprising events, autonomous goal-setting, and the pleasure of acquiring new competencies. This allows the cumulative acquisition of knowledge and skills that can later be used to accomplish fitness-enhancing goals. Intrinsic motivations continue during adulthood, and in humans artistic creativity, scientific discovery, and subjective well-being owe much to them. The study of intrinsically motivated behavior has a long history in psychological and ethological research, which is now being reinvigorated by perspectives from neuroscience, artificial intelligence and computer science. For example, recent neuroscientific research is discovering how neuromodulators like dopamine and noradrenaline relate not only to extrinsic rewards but also to novel and surprising events, how brain areas such as the superior colliculus and the hippocampus are involved in the perception and processing of events, novel stimuli, and novel associations of stimuli, and how violations of predictions and expectations influence learning and motivation. Computational approaches are characterizing the space of possible reinforcement learning algorithms and their augmentation by intrinsic reinforcements of different kinds. Research in robotics and machine learning is yielding systems with increasing autonomy and capacity for self-improvement: artificial systems with motivations that are similar to those of real organisms and support prolonged autonomous learning. Computational research on intrinsic motivation is being complemented by, and closely interacting with, research that aims to build hierarchical architectures capable of acquiring, storing, and exploiting the knowledge and skills acquired through intrinsically motivated learning. Now is an important moment in the study of intrinsically motivated open-ended development, requiring contributions and integration across a large number of fields within the cognitive sciences. This Research Topic aims to contribute to this effort by welcoming papers carried out with ethological, psychological, neuroscientific and computational approaches, as well as research that cuts across disciplines and approaches.

The two volumes LNCS 10337 and 10338 constitute the proceedings of the International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2017, held in Corunna, Spain, in June 2017. The total of 102 full papers was carefully reviewed and selected from 194 submissions during two rounds of reviewing and improvement. The papers are organized in two volumes, one on natural and artificial computation for biomedicine and neuroscience, addressing topics such as theoretical neural computation; models; natural computing in bioinformatics; physiological computing in affective smart environments; emotions; as well as signal processing and machine learning applied to biomedical and neuroscience applications. The second volume deals with biomedical applications, based on natural and artificial computing and addresses topics such as biomedical applications; mobile brain computer interaction; human robot interaction; deep learning; machine learning applied to big data analysis; computational intelligence in data coding and transmission; and applications.

The current state of the art in cognitive robotics, covering the challenges of building AI-powered intelligent robots inspired by natural cognitive systems. A novel approach to building AI-powered intelligent robots takes inspiration from the way natural cognitive systems—in humans, animals, and biological systems—develop intelligence by exploiting the full power of interactions between body and brain, the physical and social environment in which they live, and phylogenetic, developmental, and learning dynamics. This volume reports on the current state of the art in cognitive robotics, offering the first comprehensive coverage of building robots inspired by natural cognitive systems. Contributors first provide a systematic definition of cognitive robotics and a history of developments in the field. They describe in detail five main approaches: developmental, neuro, evolutionary, swarm, and soft robotics. They go on to consider methodologies and concepts, treating topics that include commonly used cognitive robotics platforms and robot simulators, biomimetic skin as an example of a hardware-based approach, machine-learning methods, and cognitive architecture. Finally, they cover the behavioral and cognitive capabilities of a variety of models, experiments, and applications, looking at issues that range from intrinsic motivation and perception to robot consciousness. Cognitive Robotics is aimed at an interdisciplinary audience, balancing technical details and examples for the computational reader with theoretical and experimental findings for the empirical scientist.

Evolution and complexity characterize both biological and artificial life – by direct modeling of biological processes and the creation of populations of interacting entities from which complex behaviors can emerge and evolve. This edited book includes invited chapters from leading scientists in the fields of artificial life, complex systems, and evolutionary computing. The contributions identify both fundamental theoretical issues and state-of-the-art real-world applications. The book is intended for researchers and graduate students in the related domains.

This volume constitutes the refereed proceedings of the 13th International Conference on Hybrid Artificial Intelligent Systems, HAIS 2018, held in Oviedo, Spain, in June 2018. The 62 full papers published in this volume were carefully reviewed and selected from 104 submissions. They are organized in the following topical sections: Neurocomputing, fuzzy systems, rough sets, evolutionary algorithms, Agents andMultiagent Systems, and alike.

Current robots and other artificial systems are typically able to accomplish only one single task. Overcoming this limitation requires the development of control architectures and learning algorithms that can support the acquisition and deployment of several different skills, which in turn seems to require a modular and hierarchical organization. In this way, different modules can acquire different skills without catastrophic interference, and higher-level components of the system can solve complex tasks by exploiting the skills encapsulated in the lower-level modules. While machine learning and robotics recognize the fundamental importance of the hierarchical organization of behavior for building robots that scale up to solve complex tasks, research in psychology and neuroscience shows increasing evidence that modularity and hierarchy are pivotal organization principles of behavior and of the brain. They might even lead to the cumulative acquisition of an ever-increasing number of skills, which seems to be a characteristic of mammals, and humans in particular. This book is a comprehensive overview of the state of the art on the modeling of the hierarchical organization of behavior in animals, and on its exploitation in robot controllers. The book perspective is highly interdisciplinary, featuring models belonging to all relevant areas, including machine learning, robotics, neural networks, and computational modeling in psychology and neuroscience. The book chapters review the authors' most recent contributions to the investigation of hierarchical behavior, and highlight the open questions and most promising research directions. As the contributing authors are among the pioneers carrying out fundamental work on this topic, the book covers the most important and topical issues in the field from a computationally informed, theoretically oriented perspective. The book will be of benefit to academic and industrial researchers and graduate students in related disciplines.