Sequential Instance-Based Learning
نویسندگان
چکیده
This paper presents and evaluates sequential instance-based learning (SIBL), an approach to action selection based upon data gleaned from prior problem solving experiences. SIBL learns to select actions based upon sequences of consecutive states. The algorithms rely primarily on sequential observations rather than a complete domain theory. We report the results of experiments on fixed-length and varying-length sequences. Four sequential similarity metrics are defined and tested: distance, convergence, consistency and recency. Model averaging and model combination methods are also tested. In the domain of three no-trump bridge play, results readily outperform IB3 on expert card selection with minimal domain knowledge.
منابع مشابه
IRDDS: Instance reduction based on Distance-based decision surface
In instance-based learning, a training set is given to a classifier for classifying new instances. In practice, not all information in the training set is useful for classifiers. Therefore, it is convenient to discard irrelevant instances from the training set. This process is known as instance reduction, which is an important task for classifiers since through this process the time for classif...
متن کاملMining Big Data: Breast Cancer Prediction using DT - SVM Hybrid Model
Breast Cancer is becoming a leading cause of death among women in the whole world; meanwhile, it is confirmed that the early detection and accurate diagnosis of this disease can ensure a long survival of the patients. This paper work presents a disease status prediction employing a hybrid methodology to forecast the changes and its consequence that is crucial for lethal infections. To alarm the...
متن کاملScatter Search for the Feature Selection Problem
The feature selection problem in the field of classification consists of obtaining a subset of variables to optimally realize the task without taking into account the remainder variables. This work presents how the search for this subset is performed using the Scatter Search metaheuristic and is compared with two traditional strategies in the literature: the Forward Sequential Selection (FSS) a...
متن کاملTransition Potential Modeling of Land-Cover based on Similarity Weighted Instance-based Learning Procedure and Its Implication in the REDD Project Design Document
Reducing Emissions from Deforestation and Forest Degradation (REDD) is a climate change mitigation strategy employed to reduce the intensity of deforestation and GHGS emissions. In recent decades, drastic land use changes in Mazandaran province caused a substantial reduction in the amount of Hyrcanian forests. The present research based on objectives of REDD projects paid to identify of fore...
متن کاملBased on Isolated Saliency or Causal Integration? Toward a Better Understanding of Human Annotation Process using Multiple Instance Learning and Sequential Probability Ratio Test
Human perception is capable of integrating local events to generate an overall impression at the global level; this is evident in daily life and is utilized repeatedly in behavioral science studies to bring objective measures into studies of human behavior. In this work, we explore two hypotheses considering whether it is the isolated-saliency or the causal-integration of information that can t...
متن کامل