Information leakage in embedding models
C Song, A Raghunathan - Proceedings of the 2020 ACM SIGSAC …, 2020 - dl.acm.org
Embeddings are functions that map raw input data to low-dimensional vector
representations, while preserving important semantic information about the inputs. Pre …
representations, while preserving important semantic information about the inputs. Pre …
Minerva: Enabling low-power, highly-accurate deep neural network accelerators
The continued success of Deep Neural Networks (DNNs) in classification tasks has sparked
a trend of accelerating their execution with specialized hardware. While published designs …
a trend of accelerating their execution with specialized hardware. While published designs …
Exponential expressivity in deep neural networks through transient chaos
We combine Riemannian geometry with the mean field theory of high dimensional chaos to
study the nature of signal propagation in deep neural networks with random weights. Our …
study the nature of signal propagation in deep neural networks with random weights. Our …
Scale-sim: Systolic cnn accelerator simulator
Systolic Arrays are one of the most popular compute substrates within Deep Learning
accelerators today, as they provide extremely high efficiency for running dense matrix …
accelerators today, as they provide extremely high efficiency for running dense matrix …
Tactics of adversarial attack on deep reinforcement learning agents
We introduce two tactics to attack agents trained by deep reinforcement learning algorithms
using adversarial examples, namely the strategically-timed attack and the enchanting attack …
using adversarial examples, namely the strategically-timed attack and the enchanting attack …
Phased lstm: Accelerating recurrent network training for long or event-based sequences
Abstract Recurrent Neural Networks (RNNs) have become the state-of-the-art choice for
extracting patterns from temporal sequences. Current RNN models are ill suited to process …
extracting patterns from temporal sequences. Current RNN models are ill suited to process …
Energy-fluctuated multiscale feature learning with deep convnet for intelligent spindle bearing fault diagnosis
X Ding, Q He - IEEE Transactions on Instrumentation and …, 2017 - ieeexplore.ieee.org
Considering various health conditions under varying operational conditions, the mining
sensitive feature from the measured signals is still a great challenge for intelligent fault …
sensitive feature from the measured signals is still a great challenge for intelligent fault …
Understanding the disharmony between dropout and batch normalization by variance shift
This paper first answers the question" why do the two most powerful techniques Dropout and
Batch Normalization (BN) often lead to a worse performance when they are combined …
Batch Normalization (BN) often lead to a worse performance when they are combined …
EESEN: End-to-end speech recognition using deep RNN models and WFST-based decoding
The performance of automatic speech recognition (ASR) has improved tremendously due to
the application of deep neural networks (DNNs). Despite this progress, building a new ASR …
the application of deep neural networks (DNNs). Despite this progress, building a new ASR …
Stripes: Bit-serial deep neural network computing
P Judd, J Albericio, T Hetherington… - 2016 49th Annual …, 2016 - ieeexplore.ieee.org
Motivated by the variance in the numerical precision requirements of Deep Neural Networks
(DNNs)[1],[2], Stripes (STR), a hardware accelerator is presented whose execution time …
(DNNs)[1],[2], Stripes (STR), a hardware accelerator is presented whose execution time …