The main objective of transfer learning is to use the knowledge acquired from a source task in order to boost the learning procedure in a target task. Transfer learning comprises a suitable solution for reinforcement learning algorithms, which often require a considerable amount of training time, especially when dealing with complex tasks. This work proposes an autonomous method for transfer le...

Eye-transfer tests, external noise manipulations, and observer models were used to systematically characterize learning mechanisms in judging motion direction of moving objects in visual periphery (Experiment 1) and fovea (Experiment 2) and to investigate the degree of transfer of the learning mechanisms from trained to untrained eyes. Perceptual learning in one eye was measured over 10 practic...

This work introduces Human-Agent Transfer (HAT), an algorithm that combines transfer learning, learning from demonstration and reinforcement learning to achieve rapid learning and high performance in complex domains. Using experiments in a simulated robot soccer domain, we show that human demonstrations transferred into a baseline policy for an agent and refined using reinforcement learning sig...

Transfer learning is about how learning from one domain or a collection of domains can be applied to another. It is learning from similarities and parallels, from experience. This paper is about a distribution free, data driven, extendable framework for transfer learning, based on the minimum description length principle. We define transfer learning in terms of a specific framework, where we ha...

The ability to transfer knowledge from one domain to another is an important aspect of learning. Knowledge transfer increases learning efficiency by freeing the learner from duplicating past efforts. In this paper, we demonstrate how reinforcement learning agents can use relational representations to transfer knowledge across related domains.

The knowledge embodied in cognitive models of smart environments, such as machine learning models, is commonly associated with time-consuming and costly processes large-scale data collection, labeling, network training, fine-tuning models. Sharing reuse these elaborated resources between intelligent systems different which known transfer learning, would facilitate the adoption services for user...

Can a neural network trained by the time series of system A be used to predict evolution B? This problem, knowing as transfer learning in broad sense, is great importance machine and data mining, yet has not been addressed for chaotic systems. Here we investigate systems from perspective synchronization-based state inference, which reservoir computer infer unmeasured variables B, while differen...