Able, III: Learning in a more visibly principled way
نویسندگان
چکیده
We describe an improved version of a cognitive model that exhibits the expert to novice transition in solving physics problems. The initial model was written by Larkin and initially translated into Soar (version 4) by Levy. In revising it to run in the latest version of Soar (7.0.0), we have updated it to be an exemplar of an understandable and reusable cognitive model: it includes graphic displays for understanding how it works, and it has had its learning mechanism organised as a general learning utility for use in other models where further principles can be specified. It is available via anonymous FTP. We argue that this example standard of displays and reusability must be more often realised if cognitive modelling is to prosper.
منابع مشابه
Able III: Learning in a more visible, principled, and reusable way
We describe an improved version of the Able cognitive model that exhibits a novice to expert transition in solving physics problems. The initial model was written by Larkin and initially translated into Soar (ver. 4) by Levy. In revising it to run in the latest version of Soar (7.0.4), we have updated it to be an exemplar of an understandable and reusable cognitive model. It includes graphic di...
متن کاملVisibly Linear Dynamic Logic
We introduce Visibly Linear Dynamic Logic (VLDL), which extends Linear Temporal Logic (LTL) by temporal operators that are guarded by visibly pushdown languages over finite words. In VLDL one can, e.g., express that a function resets a variable to its original value after its execution, even in the presence of an unbounded number of intermediate recursive calls. We prove that VLDL describes exa...
متن کاملPrincipled selection of impure measures for consistent learning of linear latent variable models
In previous work, we have developed a principled way of learning the causal structure of linear latent variable models (Silva et al., 2006). However, we have considered the case for models with pure measures only. Pure measures are observed variables that measure no more than one latent variable. This paper presents theoretical extensions that justify the selection of some types of impure measu...
متن کاملScalable Bayesian Reinforcement Learning for Multiagent POMDPs
Bayesian methods for reinforcement learning (RL) allow model uncertainty to be considered explicitly and offer a principled way of dealing with the exploration/exploitation tradeoff. However, for multiagent systems there have been few such approaches, and none of them apply to problems with state uncertainty. In this paper, we fill this gap by proposing a Bayesian RL framework for multiagent pa...
متن کاملProbabilistic Resolution of Anaphoric Reference
This paper describes the use of a Bayesian network to resolve anaphora by probabilistically combining linguistic evidence. By adopting a Bayesian approach, we are able to combine diverse evidence in a principled way, extend current understanding of linguistic phenomena by quantifying relationships empirically, and better model the non-deterministic role of linguistic evidence in resolution of a...
متن کامل