Extending Introspective Learning from Self-Models
نویسندگان
چکیده
This position paper presents open issues for using self-models to guide introspective learning, focusing on five key types of areas to explore: (1) broadening the range of learning focuses and the range of learning tools which may be brought to bear, (2) learning for self-understanding as well as self-repair, (3) making model-based approaches more sensitive to processing characteristics, instead of only outcomes, (4) making model application more flexible and robust, and (5) increasing support for self-explanation and user interaction with the meta-
منابع مشابه
A New Model of Reflective Introspective Learning
Systems which introspect about their own processes can improve their reasoning behavior in response to experience using "introspective learning" techniques. Many systems which perform introspective learning analyze and change only an underlying domain task’s reasoning processes. They do not possess the ability to r~ flectively introspect about the introspective task itself. We present a model o...
متن کاملHIRM: An Algorithm for Developing Introspective Models of Robots
We expand on previous work with the Hybrid Introspective Robot Modeling (HIRM) algorithm, which is an algorithm that develops introspective models of robots with real-world experience. These introspective models that are developed by this algorithm can then be used to accurately plan behaviors in simulation. HIRM is important because using current machine learning techniques to autonomously dev...
متن کاملIntegrated Introspective Case-Based Reasoning for Intelligent Tutoring Systems
Many intelligent tutoring systems (ITSs) have been developed, deployed, assessed, and proven to facilitate learning. However, most of these systems do not generally adapt to new circumstances, do not self-evaluate and self-configure their own strategies, and do not monitor the usage history of the learning content being delivered or presented to the students. These shortcomings force ITS develo...
متن کاملIntrospective Generative Modeling: Decide Discriminatively
We study unsupervised learning by developing introspective generative modeling (IGM) that attains a generator using progressively learned deep convolutional neural networks. The generator is itself a discriminator, capable of introspection: being able to self-evaluate the difference between its generated samples and the given training data. When followed by repeated discriminative learning, des...
متن کاملLogical Theories for Agent Introspection
Thomas Bolander: Logical Theories for Introspective Agents Artificial intelligence systems (agents) generally have models of the environments they inhabit which they use for representing facts, for reasoning about these facts and for planning actions. Much intelligent behaviour seems to involve an ability to model not only one’s external environment but also oneself and one’s own reasoning. We ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2008