The Experts below are selected from a list of 4473 Experts worldwide ranked by ideXlab platform

Jurgen Schmidhuber - One of the best experts on this subject based on the ideXlab platform.

  • deep big simple neural nets for handwritten digit recognition
    Neural Computation, 2010
    Co-Authors: Dan Ciresan, Ueli Meier, Luca Maria Gambardella, Jurgen Schmidhuber
    Abstract:

    Good old online backpropagation for plain Multilayer Perceptrons yields a very low 0.35% error rate on the MNIST handwritten digits benchmark. All we need to achieve this best result so far are many hidden layers, many neurons per layer, numerous deformed training images to avoid overfitting, and graphics cards to greatly speed up learning.

  • 2005 special issue framewise phoneme classification with bidirectional lstm and other neural network architectures
    Neural Networks, 2005
    Co-Authors: Alex Graves, Jurgen Schmidhuber
    Abstract:

    In this paper, we present bidirectional Long Short Term Memory (LSTM) networks, and a modified, full gradient version of the LSTM learning algorithm. We evaluate Bidirectional LSTM (BLSTM) and several other network architectures on the benchmark task of framewise phoneme classification, using the TIMIT database. Our main findings are that bidirectional networks outperform unidirectional ones, and Long Short Term Memory (LSTM) is much faster and also more accurate than both standard Recurrent Neural Nets (RNNs) and time-windowed Multilayer Perceptrons (MLPs). Our results support the view that contextual information is crucial to speech processing, and suggest that BLSTM is an effective architecture with which to exploit it.

  • framewise phoneme classification with bidirectional lstm and other neural network architectures
    International Joint Conference on Neural Network, 2005
    Co-Authors: Alex Graves, Jurgen Schmidhuber
    Abstract:

    In this paper, we present bidirectional Long Short Term Memory (LSTM) networks, and a modified, full gradient version of the LSTM learning algorithm. We evaluate Bidirectional LSTM (BLSTM) and several other network architectures on the benchmark task of framewise phoneme classification, using the TIMIT database. Our main findings are that bidirectional networks outperform unidirectional ones, and Long Short Term Memory (LSTM) is much faster and also more accurate than both standard Recurrent Neural Nets (RNNs) and time-windowed Multilayer Perceptrons (MLPs). Our results support the view that contextual information is crucial to speech processing, and suggest that BLSTM is an effective architecture with which to exploit it'.

Juan Moyano - One of the best experts on this subject based on the ideXlab platform.

  • optimizing the evaluation of thermal transmittance with the thermometric method using Multilayer Perceptrons
    Energy and Buildings, 2019
    Co-Authors: David Bienvenidohuertas, Carlos Rubiobellido, Juan Luis Perezordonez, Juan Moyano
    Abstract:

    Abstract The characterization of building thermal behaviour is crucial to achieve the low-carbon objectives of the European Union by 2050. In this way, the knowledge of the thermal transmittance is being developed as a significant factor of the thermophysical properties of the envelope. In the existing building, the theoretical calculation has several limitations with non-destructive techniques typical of the deterioration of the elements. Many experimental methods obtain therefore more representative results. The experimental method developed in ISO 9869–1 is the most standardized, although it presents limitations in the heat flux measurement. However, the thermometric method obtains the thermal transmittance with the surface temperature. This research is focused on the evaluation of the thermal transmittance based on ISO 9869–1 (average method and average method with correction for storage effects), but using variables measured with the thermometric method. For this purpose, Multilayer Perceptrons were used as post-processing techniques. The models were trained by using a dataset of 22,820 simulated tests of representative walls of the building stock in Spain. The determination coefficient was greater than 98% in both analysis approaches. Individual models were also generated for each building period because they significantly influenced the input variables. The results showed that thermal transmittance values can be obtained without measuring the heat flux, and the error associated with the use of tabulated values for the total internal heat transfer can be removed. This research would guarantee a high assessment tax of buildings establishing adequate energy conservation measures to improve their energy performance.

Alex Graves - One of the best experts on this subject based on the ideXlab platform.

  • 2005 special issue framewise phoneme classification with bidirectional lstm and other neural network architectures
    Neural Networks, 2005
    Co-Authors: Alex Graves, Jurgen Schmidhuber
    Abstract:

    In this paper, we present bidirectional Long Short Term Memory (LSTM) networks, and a modified, full gradient version of the LSTM learning algorithm. We evaluate Bidirectional LSTM (BLSTM) and several other network architectures on the benchmark task of framewise phoneme classification, using the TIMIT database. Our main findings are that bidirectional networks outperform unidirectional ones, and Long Short Term Memory (LSTM) is much faster and also more accurate than both standard Recurrent Neural Nets (RNNs) and time-windowed Multilayer Perceptrons (MLPs). Our results support the view that contextual information is crucial to speech processing, and suggest that BLSTM is an effective architecture with which to exploit it.

  • framewise phoneme classification with bidirectional lstm and other neural network architectures
    International Joint Conference on Neural Network, 2005
    Co-Authors: Alex Graves, Jurgen Schmidhuber
    Abstract:

    In this paper, we present bidirectional Long Short Term Memory (LSTM) networks, and a modified, full gradient version of the LSTM learning algorithm. We evaluate Bidirectional LSTM (BLSTM) and several other network architectures on the benchmark task of framewise phoneme classification, using the TIMIT database. Our main findings are that bidirectional networks outperform unidirectional ones, and Long Short Term Memory (LSTM) is much faster and also more accurate than both standard Recurrent Neural Nets (RNNs) and time-windowed Multilayer Perceptrons (MLPs). Our results support the view that contextual information is crucial to speech processing, and suggest that BLSTM is an effective architecture with which to exploit it'.

Shunichi Amari - One of the best experts on this subject based on the ideXlab platform.

Daniel S Yeung - One of the best experts on this subject based on the ideXlab platform.

  • hidden neuron pruning of Multilayer Perceptrons using a quantified sensitivity measure
    Neurocomputing, 2006
    Co-Authors: Xiaoqin Zeng, Daniel S Yeung
    Abstract:

    In the design of neural networks, how to choose the proper size of a network for a given task is an important and practical problem. One popular approach to tackling this problem is to start with an oversized network and then prune it to a smaller size so as to achieve less computational complexity and better performance in generalization. This paper presents a pruning technique, by means of a quantified sensitivity measure, to remove as many neurons as possible, those with the least relevance, from the hidden layer of a Multilayer perceptron (MLP). The sensitivity of an individual neuron is defined as the expectation of its output deviation due to expected input deviation with respect to overall inputs from a continuous interval, and the relevance of the neuron is defined as the multiplication of its sensitivity value by the summation of the absolute values of its outgoing weights. The basic idea for introducing such a relevance measure is that a neuron with less relevance ought to have less effect on its succeeding neurons and thus contribute less to the entire network. The pruning is performed by iteratively training a network to a certain performance criterion and then removing the hidden neuron with the lowest relevance value until no one can further be removed. The pruning technique is novel in its quantified sensitivity measure and so is its relevance measure. Experimental results demonstrate the effectiveness of the pruning technique.

  • hidden neuron pruning for Multilayer Perceptrons using a sensitivity measure
    International Conference on Machine Learning and Cybernetics, 2002
    Co-Authors: Daniel S Yeung, Xiaoqin Zeng
    Abstract:

    In the design of neural networks, the way to choose the proper size of a network for a given task is an important and difficult problem that still deserves further exploration. One popular approach for tackling this problem is to first use an oversized network and then pruning it to a smaller size so as to achieve less computational complexity and better performance in generalization. This paper presents a pruning technique, via a quantified sensitivity measure, to remove as many neurons as possible, those with the least relevance, from the hidden layers of a Multilayer perceptron (MLP). The sensitivity of an individual neuron is defined as the expectation of its output deviation due to expected input deviation with respect to the overall inputs from a continuous interval. The relevance of a neuron is defined as the multiplication of its sensitivity value by the summation of its outgoing weights. The basic idea is to iteratively train the network according to a certain performance criterion and then remove the neurons with the lowest relevance values. The pruning technique is novel in its quantified sensitivity measure. Computer simulations demonstrate that it works well.