The design of reward functions in reinforcement learning is a human skill that comes with experience. Unfortunately, there is not any methodology in the literature that could guide a human to design the reward function or to allow a human to transfer the skills developed in designing reward functions to another human and in a systematic manner. In this paper, we use Systematic Instructional Design, an approach in human education, to engineer a machine education methodology to design reward functions for reinforcement learning. We demonstrate the methodology in designing a hierarchical genetic reinforcement learner that adopts a neural network representation to evolve a swarm controller for an agent shepherding a boids-based swarm. The results reveal that the methodology is able to guide the design of hierarchical reinforcement learners, with each model in the hierarchy learning incrementally through a multi-part reward function. The hierarchy acts as a decision fusion function that combines the individual behaviours and skills learnt by each instruction to create a smart shepherd to control the swarm.

This paper introduces a simple and fast variant of Planet Wars as a test-bed for statistical planning based Game AI agents, and for noisy hyper-parameter optimisation. Planet Wars is a real-time strategy game with simple rules but complex game-play. The variant introduced in this paper is designed for speed to enable efficient experimentation, and also for a fixed action space to enable practical inter-operability with General Video Game AI agents. If we treat the game as a win-loss game (which is standard), then this leads to challenging noisy optimisation problems both in tuning agents to play the game, and in tuning game parameters. Here we focus on the problem of tuning an agent, and report results using the recently developed N-Tuple Bandit Evolutionary Algorithm and a number of other optimisers, including Sequential Model-based Algorithm Configuration (SMAC). Results indicate that the N-Tuple Bandit Evolutionary offers competitive performance as well as insight into the effects of combinations of parameter choices.

Two important characteristics of multi-objective evolutionary algorithms are distribution and convergency. As a classic multi-objective genetic algorithm, NSGA-II is widely used in multi-objective optimization fields. However, in NSGA-II, the random population initialization and the strategy of population maintenance based on distance cannot maintain the distribution or convergency of the population well. To dispose these two deficiencies, this paper proposes an improved algorithm, OTNSGA-II II, which has a better performance on distribution and convergency. The new algorithm adopts orthogonal experiment, which selects individuals in manner of a new discontinuing non-dominated sorting and crowding distance, to produce the initial population. And a new pruning strategy based on clustering is proposed to self-adaptively prunes individuals with similar features and poor performance in non-dominated sorting and crowding distance, or to individuals are far away from the Pareto Front according to the degree of intra-class aggregation of clustering results. The new pruning strategy makes population to converge to the Pareto Front more easily and maintain the distribution of population. OTNSGA-II and NSGA-II are compared on various types of test functions to verify the improvement of OTNSGA-II in terms of distribution and convergency.

In the last decade, several causes have determined the increasing need for rationalization of After years of economic crisis and the resulting resources within regions. The crucial ones are the reduction of resources available for research and increased globalization, mainly pursued by development investments, Smart Specialization has multinational enterprises, the economic crisis immediately become a very relevant concept to get involving all EU regions with different magnitudes these two questions answered (Foray, 2013). It and the diffusion of a new wave of general purpose represents an important chance for a progressive technologies. This situation calls for a deep economical restart. In order to develop a policyprioritization rethinking of the overall approach to regional logic to foster regional growth is development; policy-makers and experts largely important to have a deep knowledge of the potential agree on the fact that the new economic boost should evolutionary pathways related with the existing originate exploiting and enhancing the specific dynamics and the structures at regional level potential and competitive advantage of each region (McCann and Raquel Ortega-Argilès, 2013). In this through focused innovation policies. On this line, the light, each region should start this process using as European Commission has established a program standpoints the knowledge-based sectors in which labelled'Smart Specialization', consisting in a set of already presents a consistent'critical mass' or, at policies and guidelines aimed to promote the efficient and effective use of public investment in other individual to perceive.

Today's Technology world is improving day by day and therefore, we hear about numbers of Latest Technologies coming into the Tech World with much more benefits which makes our living life and our Business life much more easier than ever. Regarding this, today we are going to learn a little about AI (Artificial Intelligence). It is a Noun; the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages. What does AI (Artificial Intelligence) Mean? The modern definition of artificial intelligence (or AI) is "the study and design of intelligent agents" where an intelligent agent is a system that perceives its environment and takes actions which maximizes its chances of success.

Real-world problems often involve the optimisation of multiple conflicting objectives. These problems, referred to as multi-objective optimisation problems, are especially challenging when more than three objectives are considered simultaneously. This paper proposes an algorithm to address this class of problems. The proposed algorithm is an evolutionary algorithm based on an evolution strategy framework, and more specifically, on the Covariance Matrix Adaptation Pareto Archived Evolution Strategy (CMA-PAES). A novel selection mechanism is introduced and integrated within the framework.

Evolution strategies (ES) are a family of black-box optimization algorithms able to train deep neural networks roughly as well as Q-learning and policy gradient methods on challenging deep reinforcement learning (RL) problems, but are much faster (e.g. hours vs. days) because they parallelize better. However, many RL problems require directed exploration because they have reward functions that are sparse or deceptive (i.e. contain local optima), and it is unknown how to encourage such exploration with ES. Here we show that algorithms that have been invented to promote directed exploration in small-scale evolved neural networks via populations of exploring agents, specifically novelty search (NS) and quality diversity (QD) algorithms, can be hybridized with ES to improve its performance on sparse or deceptive deep RL tasks, while retaining scalability. Our experiments confirm that the resultant new algorithms, NS-ES and two QD algorithms, NSR-ES and NSRA-ES, avoid local optima encountered by ES to achieve higher performance on Atari and simulated robots learning to walk around a deceptive trap. This paper thus introduces a family of fast, scalable algorithms for reinforcement learning that are capable of directed exploration. It also adds this new family of exploration algorithms to the RL toolbox and raises the interesting possibility that analogous algorithms with multiple simultaneous paths of exploration might also combine well with existing RL algorithms outside ES.

Evolution strategies (ES) are a family of black-box optimization algorithms able to train deep neural networks roughly as well as Q-learning and policy gradient methods on challenging deep reinforcement learning (RL) problems, but are much faster (e.g. hours vs. days) because they parallelize better. However, many RL problems require directed exploration because they have reward functions that are sparse or deceptive (i.e. contain local optima), and it is unknown how to encourage such exploration with ES. Here we show that algorithms that have been invented to promote directed exploration in small-scale evolved neural networks via populations of exploring agents, specifically novelty search (NS) and quality diversity (QD) algorithms, can be hybridized with ES to improve its performance on sparse or deceptive deep RL tasks, while retaining scalability. Our experiments confirm that the resultant new algorithms, NS-ES and two QD algorithms, NSR-ES and NSRA-ES, avoid local optima encountered by ES to achieve higher performance on Atari and simulated robots learning to walk around a deceptive trap. This paper thus introduces a family of fast, scalable algorithms for reinforcement learning that are capable of directed exploration. It also adds this new family of exploration algorithms to the RL toolbox and raises the interesting possibility that analogous algorithms with multiple simultaneous paths of exploration might also combine well with existing RL algorithms outside ES.

Approximate computing is being considered as a promising design paradigm to overcome the energy and performance challenges in computationally demanding applications. If the case where the accuracy can be configured, the quality level versus energy efficiency or delay also may be traded-off. For this technique to be used, one needs to make sure a satisfactory user experience. This requires employing error predictors to detect unacceptable approximation errors. In this work, we propose a scheduling-aware feature selection method which leverages the intermediate results of the hardware accelerator to improve the prediction accuracy. Additionally, it configures the error predictors according to the energy consumption and latency of the system. The approach enjoys the flexibility of the prediction time for a higher accuracy. The results on various benchmarks demonstrate significant improvements in the prediction accuracy compared to the prior works which used only the accelerator inputs for the prediction.

In this paper, we propose a novel meta-learning method in a reinforcement learning setting, based on evolution strategies (ES), exploration in parameter space and deterministic policy gradients. ES methods are easy to parallelize, which is desirable for modern training architectures; however, such methods typically require a huge number of samples for effective training. We use deterministic policy gradients during adaptation and other techniques to compensate for the sample-efficiency problem while maintaining the inherent scalability of ES methods. We demonstrate that our method achieves good results compared to gradient-based meta-learning in high-dimensional control tasks in the MuJoCo simulator. In addition, because of gradient-free methods in the meta-training phase, which do not need information about gradients and policies in adaptation training, we predict and confirm our algorithm performs better in tasks that need multi-step adaptation.