Intrinsic Rewards for Maintenance, Approach, Avoidance, and Achievement Goal Types.

In reinforcement learning, reward is used to guide the learning process. The reward is often designed to be task-dependent, and it may require significant domain knowledge to design a good reward function. This paper proposes general reward functions for maintenance, approach, avoidance, and achievement goal types. These reward functions exploit the inherent property of each type of goal and are thus task-independent. We also propose metrics to measure an agent's performance for learning each type of goal. We evaluate the intrinsic reward functions in a framework that can autonomously generate goals and learn solutions to those goals using a standard reinforcement learning algorithm. We show empirically how the proposed reward functions lead to learning in a mobile robot application. Finally, using the proposed reward functions as building blocks, we demonstrate how compound reward functions, reward functions to generate sequences of tasks, can be created that allow the mobile robot to learn more complex behaviors.

Hierarchical Deep Reinforcement Learning for Continuous Action Control.

Robotic control in a continuous action space has long been a challenging topic. This is especially true when controlling robots to solve compound tasks, as both basic skills and compound skills need to be learned. In this paper, we propose a hierarchical deep reinforcement learning algorithm to learn basic skills and compound skills simultaneously. In the proposed algorithm, compound skills and basic skills are learned by two levels of hierarchy. In the first level of hierarchy, each basic skill is handled by its own actor, overseen by a shared basic critic. Then, in the second level of hierarchy, compound skills are learned by a meta critic by reusing basic skills. The proposed algorithm was evaluated on a Pioneer 3AT robot in three different navigation scenarios with fully observable tasks. The simulations were built in Gazebo 2 in a robot operating system Indigo environment. The results show that the proposed algorithm can learn both high performance basic skills and compound skills through the same learning process. The compound skills learned outperform those learned by a discrete action space deep reinforcement learning algorithm.

Assessing Human Judgment of Computationally Generated Swarming Behavior.

Computer-based swarm systems, aiming to replicate the flocking behavior of birds, were first introduced by Reynolds in 1987. In his initial work, Reynolds noted that while it was difficult to quantify the dynamics of the behavior from the model, observers of his model immediately recognized them as a representation of a natural flock. Considerable analysis has been conducted since then on quantifying the dynamics of flocking/swarming behavior. However, no systematic analysis has been conducted on human identification of swarming. In this paper, we assess subjects' assessment of the behavior of a simplified version of Reynolds' model. Factors that affect the identification of swarming are discussed and future applications of the resulting models are proposed. Differences in decision times for swarming-related questions asked during the study indicate that different brain mechanisms may be involved in different elements of the behavior assessment task. The relatively simple but finely tunable model used in this study provides a useful methodology for assessing individual human judgment of swarming behavior.

Co-Operative Coevolutionary Neural Networks for Mining Functional Association Rules.

In this paper, we introduce a novel form of association rules (ARs) that do not require discretization of continuous variables or the use of intervals in either sides of the rule. This rule form captures nonlinear relationships among variables, and provides an alternative pattern representation for mining essential relations hidden in a given data set. We refer to the new rule form as a functional AR (FAR). A new neural network-based, co-operative, coevolutionary algorithm is presented for FAR mining. The algorithm is applied to both synthetic and real-world data sets, and its performance is analyzed. The experimental results show that the proposed mining algorithm is able to discover valid and essential underlying relations in the data. Comparison experiments are also carried out with the two state-of-the-art AR mining algorithms that can handle continuous variables to demonstrate the competitive performance of the proposed method.

Trusted Autonomy and Cognitive Cyber Symbiosis: Open Challenges.

This paper considers two emerging interdisciplinary, but related topics that are likely to create tipping points in advancing the engineering and science areas. Trusted Autonomy (TA) is a field of research that focuses on understanding and designing the interaction space between two entities each of which exhibits a level of autonomy. These entities can be humans, machines, or a mix of the two. Cognitive Cyber Symbiosis (CoCyS) is a cloud that uses humans and machines for decision-making. In CoCyS, human-machine teams are viewed as a network with each node comprising humans (as computational machines) or computers. CoCyS focuses on the architecture and interface of a Trusted Autonomous System. This paper examines these two concepts and seeks to remove ambiguity by introducing formal definitions for these concepts. It then discusses open challenges for TA and CoCyS, that is, whether a team made of humans and machines can work in fluid, seamless harmony.

A game theoretic framework for incentive-based models of intrinsic motivation in artificial systems.

An emerging body of research is focusing on understanding and building artificial systems that can achieve open-ended development influenced by intrinsic motivations. In particular, research in robotics and machine learning is yielding systems and algorithms with increasing capacity for self-directed learning and autonomy. Traditional software architectures and algorithms are being augmented with intrinsic motivations to drive cumulative acquisition of knowledge and skills. Intrinsic motivations have recently been considered in reinforcement learning, active learning and supervised learning settings among others. This paper considers game theory as a novel setting for intrinsic motivation. A game theoretic framework for intrinsic motivation is formulated by introducing the concept of optimally motivating incentive as a lens through which players perceive a game. Transformations of four well-known mixed-motive games are presented to demonstrate the perceived games when players' optimally motivating incentive falls in three cases corresponding to strong power, affiliation and achievement motivation. We use agent-based simulations to demonstrate that players with different optimally motivating incentive act differently as a result of their altered perception of the game. We discuss the implications of these results both for modeling human behavior and for designing artificial agents or robots.