This paper describes a novel real-world reinforcement learning application: The Neuro Slot Car Racer. In addition to presenting the system and first results based on Neural Fitted Q-Iteration, a standard batch reinfor...
详细信息
This paper describes a novel real-world reinforcement learning application: The Neuro Slot Car Racer. In addition to presenting the system and first results based on Neural Fitted Q-Iteration, a standard batch reinforcement learning technique, an extension is proposed that is capable of improving training times and results by allowing for a reduction of samples required for successful training. The Neuralgic Pattern Selection approach achieves this by applying a failure-probability function which emphasizes neuralgic parts of the state space during sampling.
Both documents clustering and words clustering are well studied problems. Most existing algorithms cluster documents (advertisement) and words (query) separately but not simultaneously. In this paper we present a nove...
详细信息
Both documents clustering and words clustering are well studied problems. Most existing algorithms cluster documents (advertisement) and words (query) separately but not simultaneously. In this paper we present a novel idea of analyzing both queries and advertisements which occur with queries at the same time. We present an innovative co-clustering algorithm that suggests queries by co-clustering advertisements and queries. We pose the co-clustering problem as an optimization problem in information theory -the optimal co-clustering maximizes the mutual information between the clustered random variables subject to constraints on the number of row and column clusters.
AAAI was pleased to present the AAAI-08 Workshop Program, held Sunday and Monday, July 13-14, in Chicago, Illinois, USA. The program included the following 15 workshops: Advancements in POMDP Solvers;AI Education Work...
详细信息
The goal of information extraction (IE) is to find the specific information from documents composed by natural language for a particular scenario. With the development of IE methodologies, a lot of information extract...
详细信息
The goal of information extraction (IE) is to find the specific information from documents composed by natural language for a particular scenario. With the development of IE methodologies, a lot of information extraction tools have been proposed and are playing an important role in information processing. However, the efficiency of these tools may not be satisfactory to users. One of those important reasons is that most of these IE tools extract information from a single document. In this paper, we propose a extracting method which combine current single document based named extraction (NE) tool with a multi-document based radial basis function neural networks (RBFNN) for multi-document IE. The RBFNN is trained by a minimization of the localized generalization error model (L-GEM) to enhance its generalization capability. We collect a set of news pages from the Internet for the same news. Interested names are extracted by the most frequent name extracted by the NE tool. Numbers and other information that can not be extracted by NE tool will be extracted by the RBFNN by a pattern classification approach. The scenario of company layoff is used as an example to show how we extract the corresponding company name, company major location and the number of layoffs. Experimental results show the proposed method is effective and accurate.
Results are presented from an ongoing investigation testing discrimination rates of six mental tasks against the idle state for brain computer-interfacing. An online sequential classification method is employed, resul...
详细信息
Results are presented from an ongoing investigation testing discrimination rates of six mental tasks against the idle state for brain computer-interfacing. An online sequential classification method is employed, results represent calculated feedback position during trial periods. Current classification rates suggest auditory imagery shows lower discrimination against the idle state. Results mirror previous work in which linear classification accuracy was maximised within a trial window.
We consider the problem of zero-shot learning, where the goal is to learn a classifier f: X → Y that must predict novel values of Y that were omitted from the training set. To achieve this, we define the notion of a ...
详细信息
ISBN:
(纸本)9781615679119
We consider the problem of zero-shot learning, where the goal is to learn a classifier f: X → Y that must predict novel values of Y that were omitted from the training set. To achieve this, we define the notion of a semantic output code classifier (SOC) which utilizes a knowledge base of semantic properties of Y to extrapolate to novel classes. We provide a formalism for this type of classifier and study its theoretical properties in a PAC framework, showing conditions under which the classifier can accurately predict novel classes. As a case study, we build a SOC classifier for a neural decoding task and show that it can often predict words that people are thinking about from functional magnetic resonance images (fMRI) of their neural activity, even without training examples for those words.
An integral part of China's economic reforms is the privatization of state-owned enterprises (SOEs) and listing the profitable units of the SOEs on the stock market. The two stock exchanges in Shanghai and Shenzhe...
详细信息
An integral part of China's economic reforms is the privatization of state-owned enterprises (SOEs) and listing the profitable units of the SOEs on the stock market. The two stock exchanges in Shanghai and Shenzhen were opened nearly twenty years ago. The Shenzhen stock exchange market is young and energetic. Moreover, it practices a T+1 settlement rule instead of real time trade as in Hong Kong or other exchange markets. One important research question is whether there are patterns that can be identified in stock prices that can be used to develop profitable investment strategies. If strategies can be found, then this represents a violation of the efficient market hypothesis. In this work, we propose an investment strategy by using radial basis function neural networks (RBFNN) trained by localized generalization error model (L-GEM) and 4 stock price candlestick patterns. Every base RBFNN in the multiple classifier system (MCS) recognizes the occurrence of a particular candlestick pattern and the MCS combines opinions from the 4 base RBFNNs by a weighted sum to provide a final prediction. If the MCS predicts an increase for the next day, it will buy the stock and sell it within three days whenever the opening price is higher than the buy-in price or else after three days have passed. Experimental results with stocks in Shenzhen market show that our investment strategy statistically significantly outperforms a random investment, i.e. the EMH is invalid in this case.
We consider the problem of zero-shot learning, where the goal is to learn a classifier f : X → Y that must predict novel values of Y that were omitted from the training set. To achieve this, we define the notion of a...
ISBN:
(纸本)9781615679119
We consider the problem of zero-shot learning, where the goal is to learn a classifier f : X → Y that must predict novel values of Y that were omitted from the training set. To achieve this, we define the notion of a semantic output code classifier (SOC) which utilizes a knowledge base of semantic properties of Y to extrapolate to novel classes. We provide a formalism for this type of classifier and study its theoretical properties in a PAC framework, showing conditions under which the classifier can accurately predict novel classes. As a case study, we build a SOC classifier for a neural decoding task and show that it can often predict words that people are thinking about from functional magnetic resonance images (fMRI) of their neural activity, even without training examples for those words.
Visual exploratory data analysis represents a well-accepted imaging modality for high-dimensional DCE-MRI-derived breast cancer data. We employ this paradigm for discriminating between malignant and benign lesions bas...
详细信息
Visual exploratory data analysis represents a well-accepted imaging modality for high-dimensional DCE-MRI-derived breast cancer data. We employ this paradigm for discriminating between malignant and benign lesions based on different shape descriptors thanks to proven and novel dimension reduction algorithms. We demonstrate that shape structure changes such as weighted 3D Krawtchouck moments outperform global averaging moments such as geometric moment invariants in terms of discrimination of benign/malignant lesions. The best visualization of tumor shapes in a two-dimensional space is achieved based on nonlinear mapping methods, especially the ones that consider neighborhood ranks.
暂无评论