Analysis and improvement of multiple optimal learning factors for feed-forward networks

Jesudhas, Praveen; Manry, M.T.; Rawat, Rohit; Malalur, Sanjeev S.

doi:10.1109/ijcnn.2011.6033557

Cited by 6 publications

(2 citation statements)

References 36 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The derived features described in Sections 3.2 to 3.4 are computed on these data samples and then used to train a binary LSTM classifier. Stochastic learning methods such as Adam Optimizer [29] are used for training due to their faster convergence and lesser data requirement at each iteration than batch-based training algorithms [30][31][32]. The hyperparameters used to train the classifier, such as the batch size, no of LSTM units and no of epochs, are selected after observing the accuracy curves for train and test data.…”

Section: Medical Condition Detectionmentioning

confidence: 99%

A Novel Computationally Efficient Approach to Identify Visually Interpretable Medical Conditions from 2D Skeletal Data

Jesudhas¹,

Raghuveera²

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Timely identification and treatment of medical conditions could facilitate faster recovery and better health. Existing systems address this issue using custom-built sensors, which are invasive and difficult to generalize. A low-complexity scalable process is proposed to detect and identify medical conditions from 2D skeletal movements on video feed data. Minimal set of features relevant to distinguish medical conditions: AMF, PVF and GDF are derived from skeletal data on sampled frames across the entire action. The AMF (angular motion features) are derived to capture the angular motion of limbs during a specific action. The relative position of joints is represented by PVF (positional variation features). GDF (global displacement features) identifies the direction of overall skeletal movement. The discriminative capability of these features is illustrated by their variance across time for different actions. The classification of medical conditions is approached in two stages. In the first stage, a low-complexity binary LSTM classifier is trained to distinguish visual medical conditions from general human actions. As part of stage 2, a multi-class LSTM classifier is trained to identify the exact medical condition from a given set of visually interpretable medical conditions. The proposed features are extracted from the 2D skeletal data of NTU RGB + D and then used to train the binary and multi-class LSTM classifiers. The binary and multi-class classifiers observed average F1 scores of 77% and 73%, respectively, while the overall system produced an average F1 score of 69% and a weighted average F1 score of 80%. The multi-class classifier is found to utilize 10 to 100 times fewer parameters than existing 2D CNN-based models while producing similar levels of accuracy.

show abstract

Section: Medical Condition Detectionmentioning

confidence: 99%

A Novel Computationally Efficient Approach to Identify Visually Interpretable Medical Conditions from 2D Skeletal Data

Jesudhas¹,

Raghuveera²

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

show abstract

“…Such a high-level of computational complexity is mostly associated with the training sequence, which generally involves at least a single division and numerous multiplications. In this work, we use Levenberg-Marquardt backpropagation training method, which is a popular scheme for equalisation in VLC systems [11]. As for ANN equalisers, their performance is dependent on the number of taps N t and neurons N n .…”

Section: Experimental Test Setupmentioning

confidence: 99%

Visible Light Communications: Simplified Co-Equalisation of Fast OFDM in a Multiple-Input Multiple-Output Configuration

Haigh

Burton

Ghassemlooy

et al. 2019

2019 26th International Conference on Telecommunications (ICT)

View full text Add to dashboard Cite

In this paper we experimentally demonstrate, for the first time, a simplified co-equalisation for imaging multipleinput multiple-output based visible light communication systems. We show that in such systems, where all channels have similar magnitude responses, an equaliser trained on a single channel produces coefficients that may be used in the form of a look-up table to equalise the remaining channels without the need for explicit or bespoke training. The system demonstrated is based on the fast-orthogonal frequency division multiplexing based on pulse amplitude modulation scheme to improve the spectral efficiency, where a data rate of 80 Mb/s is achieved using four light-emitting diodes, each of a 4 MHz raw bandwidth. We show that the reported system performance closely matches that of the traditional and more computationally complex system in terms of bit-error rate.

show abstract

Multiple optimal learning factors for the multi-layer perceptron

2015

View full text Add to dashboard Cite

Analysis and improvement of multiple optimal learning factors for feed-forward networks

Cited by 6 publications

References 36 publications

A Novel Computationally Efficient Approach to Identify Visually Interpretable Medical Conditions from 2D Skeletal Data

A Novel Computationally Efficient Approach to Identify Visually Interpretable Medical Conditions from 2D Skeletal Data

Visible Light Communications: Simplified Co-Equalisation of Fast OFDM in a Multiple-Input Multiple-Output Configuration

Multiple optimal learning factors for the multi-layer perceptron

Contact Info

Product

Resources

About