Hardware architecture for large parallel array of Random Feature Extractors applied to image recognition

Patil, Aakash; Shen, Shanlan; Yao, Enyi; Basu, Arindam

doi:10.1016/j.neucom.2016.09.118

Cited by 20 publications

(13 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The related design approaches typically aimed efficient digital implementations of RBNs in configurable architectures. A very largescale integration (VLSI) architecture was the main target in [3,6]; the approach presented in [32] envisioned analog implementations and combined a tri-state activation function with an offline pruning procedure to limit the predictor complexity. The models proposed in [9,10,37,50] targeted FPGA implementations of the learning phase, either online or in batch mode.…”

Section: Comparison With Related Workmentioning

confidence: 99%

“…The models proposed in [9,10,37,50] targeted FPGA implementations of the learning phase, either online or in batch mode. Conversely, Decherchi et al [7] and Ragusa et al [35] proposed a minimal implementation of the forward phase of RBNs, while [32,49] introduced an effective scheme to reduce the memory requirements of the eventual predictors.…”

Section: Comparison With Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Random-based networks with dropout for embedded systems

Ragusa

Gianoglio

Zunino

et al. 2020

Neural Comput & Applic

View full text Add to dashboard Cite

Random-based learning paradigms exhibit efficient training algorithms and remarkable generalization performances. However, the computational cost of the training procedure scales with the cube of the number of hidden neurons. The paper presents a novel training procedure for random-based neural networks, which combines ensemble techniques and dropout regularization. This limits the computational complexity of the training phase without affecting classification performance significantly; the method best fits Internet of Things (IoT) applications. In the training algorithm, one first generates a pool of random neurons; then, an ensemble of independent sub-networks (each including a fraction of the original pool) is trained; finally, the sub-networks are integrated into one classifier. The experimental validation compared the proposed approach with state-of-the-art solutions, by taking into account both generalization performance and computational complexity. To verify the effectiveness in IoT applications, the training procedures were deployed on a pair of commercially available embedded devices. The results showed that the proposed approach overall improved accuracy, with a minor degradation in performance in a few cases. When considering embedded implementations as compared with conventional architectures, the speedup of the proposed method scored up to 20× in IoT devices.

show abstract

Section: Comparison With Related Workmentioning

confidence: 99%

Section: Comparison With Related Workmentioning

confidence: 99%

Random-based networks with dropout for embedded systems

Ragusa

Gianoglio

Zunino

et al. 2020

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…In [1], [18] the target was the implementation of the predictor on resource-constrained device. Hence, the underlying mapping strategy was designed to fulfill specific constraints on the admissible activation function, i.e., respectively, hardlimiter function and tristate function.…”

Section: Related Workmentioning

confidence: 99%

A Computationally Light Pruning Strategy for Single Layer Neural Networks based on Threshold Function

Ragusa

Gianoglio

Zunino

et al. 2019

2019 26th IEEE International Conference on Electronics, Circuits and Systems (ICECS)

View full text Add to dashboard Cite

Embedded machine learning relies on inference functions that can fit resource-constrained, low-power computing devices. The literature proves that single layer neural networks using threshold functions can provide a suitable trade off between classification accuracy and computational cost. In this regard, the number of neurons directly impacts both on computational complexity and on resources allocation. Thus, the present research aims at designing an efficient pruning technique that can take into account the peculiarities of the threshold function. The paper shows that feature selection criteria based on filter models can effectively be applied to neuron selection. In particular, valuable outcomes can be obtained by designing ad-hoc objective functions for the selection process. An extensive experimental campaign confirms that the proposed objective function compares favourably with state-of-the-art pruning techniques.

show abstract

“…Hence, it is best if healthy models are learnt online for each machine separately in an online manner on the deployed hardware. The choice of ELM as the classifier is motivated by both its fast convergence as well as availability of power efficient hardware [38,39] for deployment.…”

Section: Stack Of Artificial Neural Networkmentioning

confidence: 99%

A Stacked Autoencoder Neural Network based Automated Feature Extraction Method for Anomaly detection in On-line Condition Monitoring

Roy

Bose

Kar

et al. 2018

2018 IEEE Symposium Series on Computational Intelligence (SSCI)

Self Cite

View full text Add to dashboard Cite

Condition monitoring is one of the routine tasks in all major process industries. The mechanical parts such as a motor, gear, bearings are the major components of a process industry and any fault in them may cause a total shutdown of the whole process, which may result in serious losses. Therefore, it is very crucial to predict any approaching defects before its occurrence. Several methods exist for this purpose and many research are being carried out for better and efficient models. However, most of them are based on the processing of raw sensor signals, which is tedious and expensive. Recently, there has been an increase in the feature based condition monitoring, where only the useful features are extracted from the raw signals and interpreted for the prediction of the fault. Most of these are handcrafted features, where these are manually obtained based on the nature of the raw data. This of course requires the prior knowledge of the nature of data and related processes. This limits the feature extraction process. However, recent development in the autoencoder based feature extraction method provides an alternative to the traditional handcrafted approaches; however, they have mostly been confined in the area of image and audio processing. In this work, we have developed an automated feature extraction method for on-line condition monitoring based on the stack of the traditional autoencoder and an on-line sequential extreme learning machine(OSELM) network. The performance of this method is comparable to that of the traditional feature extraction approaches. The method can achieve 100% detection accuracy for determining the bearing health states of NASA bearing dataset. The simple design of this method is promising for the easy hardware implementation of Internet of Things(IoT) based prognostics solutions.

show abstract

Hardware architecture for large parallel array of Random Feature Extractors applied to image recognition

Cited by 20 publications

References 32 publications

Random-based networks with dropout for embedded systems

Random-based networks with dropout for embedded systems

A Computationally Light Pruning Strategy for Single Layer Neural Networks based on Threshold Function

A Stacked Autoencoder Neural Network based Automated Feature Extraction Method for Anomaly detection in On-line Condition Monitoring

Contact Info

Product

Resources

About