Edward Welsh’s Post

📃Scientific paper: A Hybrid EEG-based Emotion Recognition Approach Using Wavelet Convolutional Neural Networks and Support Vector Machine Abstract: INTRODUCTION: Nowadays, deep learning and convolutional neural networks (CNNs) have become widespread tools in many biomedical engineering studies. CNN is an end-to-end tool, which makes the processing procedure integrated, but in some situations, this processing tool requires to be fused with machine learning methods to be more accurate. METHODS: In this paper, a hybrid approach based on deep features extracted from wavelet CNNs (WCNNs) weighted layers and multiclass support vector machine (MSVM) was proposed to improve the recognition of emotional states from electroencephalogram (EEG) signals. First, EEG signals were preprocessed and converted to Time-Frequency (T-F) color representation or scalogram using the continuous wavelet transform (CWT) method. Then, scalograms were fed into four popular pre-trained CNNs, AlexNet, ResNet-18, VGG-19, and Inception-v3 to fine-tune them. Then, the best feature layer from each one was used as input to the MSVM method to classify four quarters of the valence-arousal model. Finally, the subject-independent leave-one-subject-out criterion was used to evaluate the proposed method on DEAP and MAHNOB-HCI databases. RESULTS: Results showed that extracting deep features from the earlier convolutional layer of ResNet-18 (Res2a) and classifying using the MSVM increased the average accuracy, precision, and recall by about 20% and 12% for MAHNOB-HCI and DEAP databases, respectively. Also, combining scalograms from four regions of pre-fro... Continued on ES/IODE ➡️ https://etcse.fr/ERnZr ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

📃Scientific paper: A Hybrid EEG-based Emotion Recognition Approach Using Wavelet Convolutional Neural Networks and Support Vector Machine Abstract: INTRODUCTION: Nowadays, deep learning and convolutional neural networks (CNNs) have become widespread tools in many biomedical engineering studies. CNN is an end-to-end tool, which makes the processing procedure integrated, but in some situations, this processing tool requires to be fused with machine learning methods to be more accurate. METHODS: In this paper, a hybrid approach based on deep features extracted from wavelet CNNs (WCNNs) weighted layers and multiclass support vector machine (MSVM) was proposed to improve the recognition of emotional states from electroencephalogram (EEG) signals. First, EEG signals were preprocessed and converted to Time-Frequency (T-F) color representation or scalogram using the continuous wavelet transform (CWT) method. Then, scalograms were fed into four popular pre-trained CNNs, AlexNet, ResNet-18, VGG-19, and Inception-v3 to fine-tune them. Then, the best feature layer from each one was used as input to the MSVM method to classify four quarters of the valence-arousal model. Finally, the subject-independent leave-one-subject-out criterion was used to evaluate the proposed method on DEAP and MAHNOB-HCI databases. RESULTS: Results showed that extracting deep features from the earlier convolutional layer of ResNet-18 (Res2a) and classifying using the MSVM increased the average accuracy, precision, and recall by about 20% and 12% for MAHNOB-HCI and DEAP databases, respectively. Also, combining scalograms from four regions of pre-fro... Continued on ES/IODE ➡️ https://etcse.fr/ERnZr ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

📃Scientific paper: A Hybrid EEG-based Emotion Recognition Approach Using Wavelet Convolutional Neural Networks and Support Vector Machine Abstract: INTRODUCTION: Nowadays, deep learning and convolutional neural networks (CNNs) have become widespread tools in many biomedical engineering studies. CNN is an end-to-end tool, which makes the processing procedure integrated, but in some situations, this processing tool requires to be fused with machine learning methods to be more accurate. METHODS: In this paper, a hybrid approach based on deep features extracted from wavelet CNNs (WCNNs) weighted layers and multiclass support vector machine (MSVM) was proposed to improve the recognition of emotional states from electroencephalogram (EEG) signals. First, EEG signals were preprocessed and converted to Time-Frequency (T-F) color representation or scalogram using the continuous wavelet transform (CWT) method. Then, scalograms were fed into four popular pre-trained CNNs, AlexNet, ResNet-18, VGG-19, and Inception-v3 to fine-tune them. Then, the best feature layer from each one was used as input to the MSVM method to classify four quarters of the valence-arousal model. Finally, the subject-independent leave-one-subject-out criterion was used to evaluate the proposed method on DEAP and MAHNOB-HCI databases. RESULTS: Results showed that extracting deep features from the earlier convolutional layer of ResNet-18 (Res2a) and classifying using the MSVM increased the average accuracy, precision, and recall by about 20% and 12% for MAHNOB-HCI and DEAP databases, respectively. Also, combining scalograms from four regions of pre-fro... Continued on ES/IODE ➡️ https://etcse.fr/ERnZr ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

📃Scientific paper: A Hybrid EEG-based Emotion Recognition Approach Using Wavelet Convolutional Neural Networks and Support Vector Machine Abstract: INTRODUCTION: Nowadays, deep learning and convolutional neural networks (CNNs) have become widespread tools in many biomedical engineering studies. CNN is an end-to-end tool, which makes the processing procedure integrated, but in some situations, this processing tool requires to be fused with machine learning methods to be more accurate. METHODS: In this paper, a hybrid approach based on deep features extracted from wavelet CNNs (WCNNs) weighted layers and multiclass support vector machine (MSVM) was proposed to improve the recognition of emotional states from electroencephalogram (EEG) signals. First, EEG signals were preprocessed and converted to Time-Frequency (T-F) color representation or scalogram using the continuous wavelet transform (CWT) method. Then, scalograms were fed into four popular pre-trained CNNs, AlexNet, ResNet-18, VGG-19, and Inception-v3 to fine-tune them. Then, the best feature layer from each one was used as input to the MSVM method to classify four quarters of the valence-arousal model. Finally, the subject-independent leave-one-subject-out criterion was used to evaluate the proposed method on DEAP and MAHNOB-HCI databases. RESULTS: Results showed that extracting deep features from the earlier convolutional layer of ResNet-18 (Res2a) and classifying using the MSVM increased the average accuracy, precision, and recall by about 20% and 12% for MAHNOB-HCI and DEAP databases, respectively. Also, combining scalograms from four regions of pre-fro... Continued on ES/IODE ➡️ https://etcse.fr/ERnZr ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

At the end of 1993 I defended my PhD thesis titled "Connections, Neurons and Activation. The organisation of representation in artificial neural networks". This was 30 years ago and well before deep learning arrived but after the introduction of back propagration that inspired me to start my PhD research in 1988. Most of the topics are outdated but some are still relevant and or interesting. For instance my interpretation as an ingeneer on how neural networks work in the brain (sections 1.1 and 3.1), using complex values in neural networks and how this effects back propagation and gradient descent (chapter 2, this may still be interesting for positional encoding in transformers), the striking resemblence between computational maps in the brain (e.g. somatosensory cortex) and dimension reduction in Kohonen self-organizing feature maps and how to represent information in temporal-actitivity patterns in a vibrating membrane model. Throughout the book I use tensor notation to simplify the mathematics but unfornately we didn't have any Cuda or TPU back then. So if you wonder why it's called tensorflow, have a look at the mathematics ;-) Thanks again H. Jaap van den Herik Eric Postma and Peter J. Braspenning , it was great 30 years ago but it is getting even better now!

📃Scientific paper: A Hybrid EEG-based Emotion Recognition Approach Using Wavelet Convolutional Neural Networks and Support Vector Machine Abstract: INTRODUCTION: Nowadays, deep learning and convolutional neural networks (CNNs) have become widespread tools in many biomedical engineering studies. CNN is an end-to-end tool, which makes the processing procedure integrated, but in some situations, this processing tool requires to be fused with machine learning methods to be more accurate. METHODS: In this paper, a hybrid approach based on deep features extracted from wavelet CNNs (WCNNs) weighted layers and multiclass support vector machine (MSVM) was proposed to improve the recognition of emotional states from electroencephalogram (EEG) signals. First, EEG signals were preprocessed and converted to Time-Frequency (T-F) color representation or scalogram using the continuous wavelet transform (CWT) method. Then, scalograms were fed into four popular pre-trained CNNs, AlexNet, ResNet-18, VGG-19, and Inception-v3 to fine-tune them. Then, the best feature layer from each one was used as input to the MSVM method to classify four quarters of the valence-arousal model. Finally, the subject-independent leave-one-subject-out criterion was used to evaluate the proposed method on DEAP and MAHNOB-HCI databases. RESULTS: Results showed that extracting deep features from the earlier convolutional layer of ResNet-18 (Res2a) and classifying using the MSVM increased the average accuracy, precision, and recall by about 20% and 12% for MAHNOB-HCI and DEAP databases, respectively. Also, combining scalograms from four regions of pre-fro... Continued on ES/IODE ➡️ https://etcse.fr/ERnZr ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

MIT graduate students have made a breakthrough discovery in the field of neural networks. For the first time, they have demonstrated that trained neural networks have a similar underlying representation as the human auditory system - specifically when trained with sound in noise. This pivotal paper provides a new understanding of human hearing and challenges the hypothesis that statistical methods alone cannot capture it. Learn more by reading the paper at the link below. Link: https://lnkd.in/e6_A2GYV

Trustworthy AI - Test adequacy for Deep Neural Networks (DNNs) How do you know if you have sufficiently tested a DNN? Is your test suite sufficient, regardless of how you built it? Can you trust your DNN's high accuracy results? We have addressed these questions by proposing a methodology (TEASMA) and performing a massive empirical study. Results show that post-training mutation yields more accurate results but other metrics are less computationally expensive and can be reasonable alternatives. The following paper was accepted in IEEE Transactions on Software Engineering.  “TEASMA: A Practical Methodology for Test Adequacy Assessment of Deep Neural Networks”, Amin Abbasi, Mahboubeh Dadkhah, PhD, Lionel Briand, Dayi Lin, Ph.D. This is work is part of Amin Abbasi's PhD (University of Ottawa) and Mahboubeh Dadkhah's postdoc, as well as a research collaboration with Huawei Canada. Preprint:

"National University of Singapore (NUS) researchers have demonstrated that deep learning allows them to observe the dynamics of single molecules more precisely and with less data than traditional evaluation methods. They used convolutional neural networks (CNNs) to observe the movement of single molecules in artificial systems, cells and small organisms. Their findings have been published in the Biophysical Journal. This method promises to accelerate single-molecule measurements in complex systems and make it more accessible to a wider range of researchers. A single molecule is the most basic unit observable in biological systems. Understanding its behavior and interactions unlocks insights into the functioning of biological systems, paving the way for strategic interventions in diseases." #dl #imagingtechnology

New research article Slimmed Optical Neural Networks with Multiplexed Neuron Sets and a Corresponding Backpropagation Training Algorithm https://lnkd.in/gN_5v47W Yi-Feng Liu, Rui-Yao Ren, Dai-Bao Hou, Hai-Zhong Weng, Bo-Wen Wang, Ke-Jie Huang, Xing Lin, Feng Liu, Chen-Hui Li and Chaoyuan Jin Optical neural networks (ONNs) have recently attracted extensive interest as potential alternatives to electronic artificial neural networks, owing to their intrinsic capabilities in parallel signal processing with reduced power consumption and low latency. Preliminary confirmation of parallelism in optical computing has been widely performed by applying wavelength division multiplexing (WDM) to the linear transformation of neural networks. However, interchannel crosstalk has obstructed WDM technologies from being deployed in nonlinear activation on ONNs. Here, we propose a universal WDM structure called multiplexed neuron sets (MNS), which applies WDM technologies to optical neurons and enables ONNs to be further compressed. A corresponding backpropagation (BP) training algorithm was proposed to alleviate or even annul the influence of interchannel crosstalk in MNS-based WDM-ONNs. For simplicity, semiconductor optical amplifiers are employed as an example of MNS to construct a WDM-ONN trained using the new algorithm. The results show that the combination of MNS and the corresponding BP training algorithm clearly downsizes the system and improves the energy efficiency by a factor of 10 while providing similar performance to traditional ONNs.