The Experts below are selected from a list of 20763 Experts worldwide ranked by ideXlab platform

Anthony S Maida - One of the best experts on this subject based on the ideXlab platform.

  • multi Layer unsupervised learning in a spiking convolutional neural network
    International Joint Conference on Neural Network, 2017
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Spiking neural networks (SNNs) have advantages over traditional, non-spiking networks with respect to biorealism, potential for low-power hardware implementations, and theoretical computing power. However, in practice, spiking networks with multi-Layer learning have proven difficult to train. This paper explores a novel, bio-inspired spiking convolutional neural network (CNN) that is trained in a greedy, Layer-wise fashion. The spiking CNN consists of a convolutional/pooling Layer followed by a feature Discovery Layer, both of which undergo bio-inspired learning. Kernels for the convolutional Layer are trained using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer uses a probabilistic spike-timing-dependent plasticity (STDP) learning rule. This Layer represents complex visual features using WTA-thresholded, leaky, integrate-and-fire (LIF) neurons. The new model is evaluated on the MNIST digit dataset using clean and noisy images. Intermediate results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning architecture. The recognition performance for clean images is above 98%. This performance is accounted for by the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5%. This level of performance loss indicates that the network is robust to additive noise.

  • IJCNN - Multi-Layer unsupervised learning in a spiking convolutional neural network
    2017 International Joint Conference on Neural Networks (IJCNN), 2017
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Spiking neural networks (SNNs) have advantages over traditional, non-spiking networks with respect to biorealism, potential for low-power hardware implementations, and theoretical computing power. However, in practice, spiking networks with multi-Layer learning have proven difficult to train. This paper explores a novel, bio-inspired spiking convolutional neural network (CNN) that is trained in a greedy, Layer-wise fashion. The spiking CNN consists of a convolutional/pooling Layer followed by a feature Discovery Layer, both of which undergo bio-inspired learning. Kernels for the convolutional Layer are trained using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer uses a probabilistic spike-timing-dependent plasticity (STDP) learning rule. This Layer represents complex visual features using WTA-thresholded, leaky, integrate-and-fire (LIF) neurons. The new model is evaluated on the MNIST digit dataset using clean and noisy images. Intermediate results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning architecture. The recognition performance for clean images is above 98%. This performance is accounted for by the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5%. This level of performance loss indicates that the network is robust to additive noise.

  • bio inspired spiking convolutional neural network using Layer wise sparse coding and stdp learning
    arXiv: Neural and Evolutionary Computing, 2016
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Hierarchical feature Discovery using non-spiking convolutional neural networks (CNNs) has attracted much recent interest in machine learning and computer vision. However, it is still not well understood how to create a biologically plausible network of brain-like, spiking neurons with multi-Layer, unsupervised learning. This paper explores a novel bio-inspired spiking CNN that is trained in a greedy, Layer-wise fashion. The proposed network consists of a spiking convolutional-pooling Layer followed by a feature Discovery Layer extracting independent visual features. Kernels for the convolutional Layer are trained using local learning. The learning is implemented using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer extracts independent features by probabilistic, leaky integrate-and-fire (LIF) neurons that are sparsely active in response to stimuli. The Layer of the probabilistic, LIF neurons implicitly provides lateral inhibitions to extract sparse and independent features. Experimental results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning. The visual features obtained from the proposed probabilistic LIF neurons in the feature Discovery Layer are utilized for training a classifier. Classification results contribute to the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The proposed model is evaluated on the MNIST digit dataset using clean and noisy images. The recognition performance for clean images is above 98%. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5% depending on noise types and densities. This level of performance loss indicates that the network is robust to additive noise.

Amirhossein Tavanaei - One of the best experts on this subject based on the ideXlab platform.

  • multi Layer unsupervised learning in a spiking convolutional neural network
    International Joint Conference on Neural Network, 2017
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Spiking neural networks (SNNs) have advantages over traditional, non-spiking networks with respect to biorealism, potential for low-power hardware implementations, and theoretical computing power. However, in practice, spiking networks with multi-Layer learning have proven difficult to train. This paper explores a novel, bio-inspired spiking convolutional neural network (CNN) that is trained in a greedy, Layer-wise fashion. The spiking CNN consists of a convolutional/pooling Layer followed by a feature Discovery Layer, both of which undergo bio-inspired learning. Kernels for the convolutional Layer are trained using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer uses a probabilistic spike-timing-dependent plasticity (STDP) learning rule. This Layer represents complex visual features using WTA-thresholded, leaky, integrate-and-fire (LIF) neurons. The new model is evaluated on the MNIST digit dataset using clean and noisy images. Intermediate results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning architecture. The recognition performance for clean images is above 98%. This performance is accounted for by the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5%. This level of performance loss indicates that the network is robust to additive noise.

  • IJCNN - Multi-Layer unsupervised learning in a spiking convolutional neural network
    2017 International Joint Conference on Neural Networks (IJCNN), 2017
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Spiking neural networks (SNNs) have advantages over traditional, non-spiking networks with respect to biorealism, potential for low-power hardware implementations, and theoretical computing power. However, in practice, spiking networks with multi-Layer learning have proven difficult to train. This paper explores a novel, bio-inspired spiking convolutional neural network (CNN) that is trained in a greedy, Layer-wise fashion. The spiking CNN consists of a convolutional/pooling Layer followed by a feature Discovery Layer, both of which undergo bio-inspired learning. Kernels for the convolutional Layer are trained using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer uses a probabilistic spike-timing-dependent plasticity (STDP) learning rule. This Layer represents complex visual features using WTA-thresholded, leaky, integrate-and-fire (LIF) neurons. The new model is evaluated on the MNIST digit dataset using clean and noisy images. Intermediate results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning architecture. The recognition performance for clean images is above 98%. This performance is accounted for by the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5%. This level of performance loss indicates that the network is robust to additive noise.

  • bio inspired spiking convolutional neural network using Layer wise sparse coding and stdp learning
    arXiv: Neural and Evolutionary Computing, 2016
    Co-Authors: Amirhossein Tavanaei, Anthony S Maida
    Abstract:

    Hierarchical feature Discovery using non-spiking convolutional neural networks (CNNs) has attracted much recent interest in machine learning and computer vision. However, it is still not well understood how to create a biologically plausible network of brain-like, spiking neurons with multi-Layer, unsupervised learning. This paper explores a novel bio-inspired spiking CNN that is trained in a greedy, Layer-wise fashion. The proposed network consists of a spiking convolutional-pooling Layer followed by a feature Discovery Layer extracting independent visual features. Kernels for the convolutional Layer are trained using local learning. The learning is implemented using a sparse, spiking auto-encoder representing primary visual features. The feature Discovery Layer extracts independent features by probabilistic, leaky integrate-and-fire (LIF) neurons that are sparsely active in response to stimuli. The Layer of the probabilistic, LIF neurons implicitly provides lateral inhibitions to extract sparse and independent features. Experimental results show that the convolutional Layer is stack-admissible, enabling it to support a multi-Layer learning. The visual features obtained from the proposed probabilistic LIF neurons in the feature Discovery Layer are utilized for training a classifier. Classification results contribute to the independent and informative visual features extracted in a hierarchy of convolutional and feature Discovery Layers. The proposed model is evaluated on the MNIST digit dataset using clean and noisy images. The recognition performance for clean images is above 98%. The performance loss for recognizing the noisy images is in the range 0.1% to 8.5% depending on noise types and densities. This level of performance loss indicates that the network is robust to additive noise.

Courtney Greene - One of the best experts on this subject based on the ideXlab platform.

  • The Search for a New OPAC: Selecting an Open Source Discovery Layer
    Serials Review, 2012
    Co-Authors: Kate B. Moore, Courtney Greene
    Abstract:

    In early 2011, an Indiana University Libraries task force was charged with selecting an open source Discovery Layer to serve as the public interface for IU's online catalog, IUCAT. This process included creating a rubric of core functionality and rating two Discovery Layers based on criteria in four main categories: general features and functionality; authentication and account management; export and share; and search functionality and results display. The article includes information about our rubric and the two Discovery Layers reviewed, Blacklight and VuFind, as well as a discussion of the priorities of the task force. The article concludes with future steps and anticipated highlights for IUCAT.

  • Choosing Discovery: A Literature Review on the Selection and Evaluation of Discovery Layers
    Journal of Web Librarianship, 2012
    Co-Authors: Kate B. Moore, Courtney Greene
    Abstract:

    Within the next few years, traditional online public access catalogs will be replaced by more robust and interconnected Discovery Layers that can serve as primary public interfaces to simultaneously search many separate collections of resources. Librarians have envisioned this type of Discovery tool since the 1980s, and research shows that Discovery Layer functionality and features have already resolved, or will soon remedy, many of the known issues with the traditional online public access catalog. The authors of this article review the literature on Discovery, focusing on the evolution from the traditional online public access catalog to newer Discovery interfaces, and summarize what has been published regarding the selection and evaluation of these new tools. Based on this review, emerging trends in the acquisition of Discovery Layers are described, including: the amount of time libraries devote to selection and evaluation, the staff involved and their areas of expertise, which Discovery tools were sel...

Wendy Camber - One of the best experts on this subject based on the ideXlab platform.

Jessica Morales - One of the best experts on this subject based on the ideXlab platform.

  • but i just want a book is your Discovery Layer meeting your users needs
    Journal of Web Librarianship, 2018
    Co-Authors: Christine Rigda, Margaret A Hoogland, Jessica Morales
    Abstract:

    In January 2016, the University of Toledo Libraries implemented EBSCO Discovery Services (EDS) as its Discovery Layer. Administrators questioned whether users were able to find consortial material ...

  • “But I just want a book!” Is your Discovery Layer meeting your users’ needs?
    Journal of Web Librarianship, 2018
    Co-Authors: Christine Rigda, Margaret A Hoogland, Jessica Morales
    Abstract:

    AbstractIn January 2016, the University of Toledo Libraries implemented EBSCO Discovery Services (EDS) as its Discovery Layer. Administrators questioned whether users were able to find consortial material in the EDS, so they assembled a task force to conduct a pilot usability test. The task force gathered demographic data and recorded the screens of 25 students answering six task questions. Results showed participants could easily find most items except books, and for tasks that were open-ended, many students continued searching even though they found relevant material. To determine why participants could not find books, the task force consulted with EBSCO and discovered a configuration problem that was easily resolved by editing a mapping table and adding a custom limiter for print books. The searching issue was more difficult to determine, and the task force suggests a lack of library instruction may be at least partly to blame. Libraries invest significant resources in Discovery Layers. If users have d...