Abstract In this paper we propose a method to augment Reinforcement Learning agent with legibility. This is inspired by the literature in Explainable Planning and allows regularize agent’s policy after training, without requiring modify its learning algorithm. achieved evaluating how optimal may produce observations that would make an observer model infer wrong policy. our formulation, decision...

The research field concerned with the digital restoration of degraded written heritage lacks a quantitative metric for evaluating its results, which prevents comparison relevant methods on large datasets. Thus, we introduce novel dataset Subjective Assessments Legibility in Ancient Manuscript Images (SALAMI) to serve as ground truth development evaluation metrics text restoration. This consists...

Executive Summary of Findings This document contains review of literature on the effects of vibration on human performance. It is intended to address the issues relevant to reported difficulties in the use of the XM4 Command and Control Vehicle while the vehicle is in motion. The focus of this review is on the effects of vibration on visual task, with additional consideration given to manual ta...

We present a novel method for automatically enhancing (or diminishing) the legibility of typefaces without altering grapheme size or energy. Just as caricature generators enhance those features which most distinguish a targeted face from the average face, we generate an enhanced grapheme by exaggerating those aspects which distinguish that grapheme from an average grapheme. More particularly, w...