Active Feature Acquisition with POMDP Models
نویسندگان
چکیده
We consider the problem of active feature acquisition (AFA), where the selection of a new feature is conditional on the instantiations of previously selected features. The problem is formulated as a partially observable Markov decision process (POMDP). We present a method to construct an approximate POMDP for the AFA problem and discuss its accuracy. We propose a non-stationary policy to improve the classification performance at lower feature acquisition costs. The methods are demonstrated on medical diagnosis problems.
منابع مشابه
Developing a Feature Extraction of Existing Structures Using an Ambient Vibration Test
The paper aims to extract the dynamic properties of existing structures without utilizing the analytical models. The ambient vibration testing could be used on any type of frame such as concrete, steel and masonry to investigate the structural vulnerability. The method could be the first stage and necessarily for the retrofit process. To achieve this aim, the ambient vibration testing can...
متن کاملActive Feature Acquisition for Classifier Induction
Many induction problems, such as on-line customer profiling, include missing data that can be acquired at a cost, such as incomplete customer information that can be filled in by an intermediary. For building accurate predictive models, acquiring complete information for all instances is often prohibitively expensive or unnecessary. Randomly selecting instances for feature acquisition allows a ...
متن کاملThe Acquisition of Definiteness Feature by Persian L2 Learners of English
The definiteness feature in English is both LF and PF interpretable while Persian is a language in which this feature is LF-interpretable but PF-uninterpretable. Hence, there is no overt article or morphological inflection in Persian denoting a definite context. Furthermore, Persian partially encodes specificity not definiteness. In definiteness both the speaker and hearer are involved while in...
متن کاملBreaking Boundaries: Active Information Acquisition Across Learning and Diagnosis
To date, the processes employed for active information acquisition during periods of learning and diagnosis have been considered as separate and have been applied in distinct phases of analysis. While active learning centers on the collection of information about training cases in order to build better predictive models, diagnosis uses fixed predictive models for guiding the collection of obser...
متن کاملA POMDP based approach to optimally select sellers in electronic marketplaces
Selecting a seller in e-markets is a tedious task that we might want to delegate to an agent. Many approaches to constructing such agents have been proposed, building upon different foundations (decision theory, trust modeling) and making use of different information (direct experience with sellers, reputation of sellers, trustworthiness of other buyers called advisors, etc.). In this paper, we...
متن کامل