Skill Rating by Bayesian Inference

Systems Engineering often involves computer modelling the behaviour of proposed systems and their components. Where a component is human, fallibility must be modelled by a stochastic agent. The identification of a model of decision-making over quantifiable options is investigated using the game-domain of Chess. Bayesian methods are used to infer the distribution of players’ skill levels from the moves they play rather than from their competitive results. The approach is used on large sets of games by players across a broad FIDE Elo range, and is in principle applicable to any scenario where high-value decisions are being made under pressure.

Borg's rating of perceived exertion scales: do the verbal anchors mean the same for different clinical groups?

Objective: To examine the interpretation of the verbal anchors used in the Borg rating of perceived exertion (RPE) scales in different clinical groups and a healthy control group. Design: Prospective experimental study. Setting: Rehabilitation center. Participants: Nineteen subjects with brain injury, 16 with chronic low back pain (CLBP), and 20 healthy controls. Interventions: Not applicable. Main Outcome Measures: Subjects used a visual analog scale (VAS) to rate their interpretation of the verbal anchors from the Borg RPE 6-20 and the newer 10-point category ratio scale. Results: All groups placed the verbal anchors in the order that they occur on the scales. There were significant within-group differences (P > .05) between VAS scores for 4 verbal anchors in the control group, 8 in the CLBP group, and 2 in the brain injury group. There was no significant difference in rating of each verbal anchor between the groups (P > .05). Conclusions: All subjects rated the verbal anchors in the order they occur on the scales, but there was less agreement in rating of each verbal anchor among subjects in the brain injury group. Clinicians should consider the possibility of small discrepancies in the meaning of the verbal anchors to subjects, particularly those recovering from brain injury, when they evaluate exercise perceptions.

Intrinsic chess rating

This paper develops and tests formulas for representing playing strength at chess by the quality of moves played, rather than by the results of games. Intrinsic quality is estimated via evaluations given by computer chess programs run to high depth, ideally so that their playing strength is sufficiently far ahead of the best human players as to be a `relatively omniscient' guide. Several formulas, each having intrinsic skill parameters s for `sensitivity' and c for `consistency', are argued theoretically and tested by regression on large sets of tournament games played by humans of varying strength as measured by the internationally standard Elo rating system. This establishes a correspondence between Elo rating and the parameters. A smooth correspondence is shown between statistical results and the century points on the Elo scale, and ratings are shown to have stayed quite constant over time. That is, there has been little or no `rating inflation'. The theory and empirical results are transferable to other rational-choice settings in which the alternatives have well-defined utilities, but in which complexity and bounded information constrain the perception of the utility values.