Skip to search formSkip to main contentSkip to account menu
Fixed Point Quantization of Deep Convolutional Networks
TLDR
This paper proposes a quantizer design for fixed point implementation of DCNs, formulate and solve an optimization problem to identify optimal fixed point bit-width allocation across DCN layers, and demonstrates that fine-tuning can further enhance the accuracy of fixed point DCNs beyond that of the original floating point model.
View PDF on arXiv
Improving performance of recurrent neural network with relu nonlinearity
TLDR
This paper offers a simple dynamical systems perspective on weight initialization process, which allows for a modified weight initialization strategy, and shows that this initialization technique leads to successfully training RNNs composed of ReLUs.
View PDF on arXiv
OpenEDS: Open Eye Dataset
TLDR
It is anticipated that OpenEDS will create opportunities to researchers in the eye tracking community and the broader machine learning and computer vision community to advance the state of eye-tracking for VR applications.
View PDF on arXiv
Deep Recurrent Neural Networks for seizure detection and early seizure detection systems
TLDR
A deep learning framework via the use of Gated Recurrent Unit (GRU) RNNs for seizure detection and early seizure detection systems is proposed, which can detect about 98% of seizure events within the first 5 seconds of the overall epileptic seizure duration.
View PDF on arXiv
Support vector machines for seizure detection in an animal model of chronic epilepsy
TLDR
It is found that SVDD not only performed better than the other SVM types in terms of highest value of the mean optimality index metric () but also gave a more reliable performance across the two EEG datasets.
View on IOP Publishing
Fast SVM training using approximate extreme points
TLDR
The AESVM implementation was found to train much faster than the other methods, while its classification accuracy was similar to that of LIBSVM in all cases, and it gave competitively fast classification times.
View PDF on arXiv
Overcoming Challenges in Fixed Point Training of Deep Convolutional Networks
TLDR
This work is an attempt to draw a theoretical connection between low numerical precision and training algorithm stability, and propose and verify through experiments methods that are able to improve the training performance of deep convolutional networks in fixed point.
View PDF on arXiv
Epilepsy Detection and Monitoring
TLDR
This chapter investigates how wavelets, synchronization, Lyapunov exponents, principal component analysis, and other techniques can help investigators extract information about impending seizures and illustrates how these techniques can be brought together in a closed-loop seizure prevention system.
OpenEDS2020: Open Eyes Dataset
We present the second edition of OpenEDS dataset, OpenEDS2020, a novel dataset of eye-image sequences captured at a frame rate of 100 Hz under controlled illumination, using a virtual-reality…
View PDF on arXiv
Computational Modeling of Channelrhodopsin-2 Photocurrent Characteristics in Relation to Neural Signaling
TLDR
This paper analytically investigates the conditions under which the characteristic specific to the 3-state model, namely the monoexponential photocurrent decay of the newly developed variants of ChR2, can occur in the framework of the 4- state model.
View on Springer
...
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy, Terms of Service, and Dataset License