23 papers · ranked by Valyu relevance
Jimmy Lin
In this work, we observe while gradient compression generally has a negative impact on test accuracy in non-private training, it can sometimes improve test accuracy in differentially private training. Specifically, we observe that when employing aggressive sparsification or rank reduction to the gradients, test…
Shuer Ye, Jasper Christian Mostert, Robin Pedersen, Xiaqing Lan + 3 more
The functional cortical hierarchy of the human brain, a fundamental principle of brain organization, has been extensively characterized during resting state for healthy younger adults. However, functional re-organization during naturalistic settings, such as movie-watching, and its alterations across the adult lifespan…
Hui‐Po Wang, Mario Fritz
General Neural Parameter Prior Models Authors: ['Hui‐Po Wang' 'Mario Fritz'] Despite the widespread use of statistical prior models in various fields, such models for neural network gradients have long been overlooked. The inherent challenge stems from their high-dimensional structures and complex interdependencies…
Sian Jin, Chengming Zhang, Xintong Jiang, Yunhe Feng + 4 more
Using Error-Bounded Lossy Compression Authors: ['Sian Jin' 'Chengming Zhang' 'Xintong Jiang' 'Yunhe Feng' 'Hui Guan' 'Guanpeng Li' 'Shuaiwen Leon Song' 'Dingwen Tao'] Deep neural networks (DNNs) are becoming increasingly deeper, wider, and non-linear due to the growing demands on prediction accuracy and analysis…
Hexiang Wu, Jia Qu, Linzhi Wu, Jianxun Zhang + 1 more
The impact mechanical properties of graded hourglass lattice sandwich structures under impact compression were studied using experiments and numerical simulations. The influence of the gradient distribution on the deformation mode, peak load, and energy absorption capacity of the hourglass lattice sandwich structure…
Xinmei Xiang, Rujin Wang, Jianzhang Huang, Jiale Huang + 2 more
This study investigates the quasi-static equalbiaxial compression behavior of 3D-printed Miura-ori metamaterials incorporating both out-of-plane and in-plane gradient configurations. The Miura-ori structure, composed of tessellated parallelogram units, exhibits pronounced anisotropic behavior due to its unique folding…
Yun Zhai, Tianyuan Zhong, Shuangquan Guo, Sheng Lin + 3 more
'Xiaowei Ma' 'Dan Sameoto'] Porous bone implants have been extensively studied, with gradient structures receiving increasing attention due to their superior compatibility with bone tissue. However, comparative studies between gradient and uniform structures remain relatively scarce. In this study, selective laser…
Takuya Ito, John D. Murray
Human cognition recruits diverse neural processes, yet the organizing computational and functional architectures remain unclear. Here, we characterized the geometry and topography of multi-task representations across human cortex using functional MRI during 26 cognitive tasks in the same subjects. We measured the…
Zecheng Li, Yifei Ren, Sharvari Kemkar, Paul Mollenkopf + 5 more
The mechanical properties of the extracellular matrix (ECM) regulate tumor growth and invasion in the tumor microenvironment. Models of biopolymer networks have been used to investigate the impact of elasticity and viscoelasticity of ECM on tumor behavior. Under tumor compression, these networks also show poroelastic…
Defu Liu, Yixiao Zhu, Zhe Liu, Yi Liu + 4 more
'Ruihao Li' 'Wei Yi'] The exceptional performance of general-purpose large models has driven various industries to focus on developing domain-specific models. However, large models are not only time-consuming and labor-intensive during the training phase but also have very high hardware requirements during the…
Yanlong Gao, Haiming Xu, Wei Huang, Hao Bai + 3 more
With the extensive applications of satellite image data in environmental monitoring and geographic surveying and mapping, the amount of data has increased rapidly, which brings great challenges for transmitting and storing these images. However, when processing high-resolution and multi-spectral satellite data…
Felix Zeller, Chieh-Min Hsieh, Wilke Dononelli, Tim Neudecker
The field of liquid-phase and solid-state high-pressure chemistry has exploded since the advent of the diamond anvil cell, an experimental technique that allows the application of pressures up to several hundred gigapascal. To complement high-pressure experiments, a large number of computational tools have been…
David Podgorelec, Damjan Strnad, Ivana Kolingerová, Borut Žalik + 1 more
'Jun Chen'] After a boom that coincided with the advent of the internet, digital cameras, digital video and audio storage and playback devices, the research on data compression has rested on its laurels for a quarter of a century. Domain-dependent lossy algorithms of the time, such as JPEG, AVC, MP3 and others…
Luke Staniscia, Yun William Yu
Because of the rapid generation of data, the study of compression algorithms to reduce storage and transmission costs is important to bioinformaticians. Much of the focus has been on sequence data, including both genomes and protein amino acid sequences stored in FASTA files. Current standard practice is to use an…
Yibo Yang, Stephan Mandt, Lucas Theis
Neural compression is the application of neural networks and other machine learning methods to data compression. Recent advances in statistical machine learning have opened up new possibilities for data compression, allowing compression algorithms to be learned end-to-end from data using powerful generative models such…
Senik Matinyan, Jan Pieter Abrahams
High-throughput data collection in crystallography poses significant challenges in handling massive amounts of data. Here, we present TERSE, a novel lossless compression algorithm specifically designed for diffraction data. We compare TERSE with the established lossless compression algorithms implemented in gzip, CBF…
Christian D. Rask, Daniel E. Lucani
Embedded Applications Authors: ['Christian D. Rask' 'Daniel E. Lucani'] We introduce RAGE, an image compression framework that achieves four generally conflicting objectives: 1) good compression for a wide variety of color images, 2) computationally efficient, fast decompression, 3) fast random access of images with…
Authors not listed
This work provides a rigorous theoretical investigation of selective error correction strategies for variational quantum algorithms, with focus on understanding the interplay between error suppression, circuit trainability, and computational resource requirements. We develop a mathematical framework that characterizes…
Mateen Ulhaq
| Name: | Mateen Ulhaq | | --- | --- | | Degree: | Master of Applied Science (Engineering) | | Thesis title: | Learned Compression for Images and Point Clouds | | Committee: | Chair: Andrew Rawicz | | | Professor, Engineering Science | | | Ivan V. Bajić | | | Supervisor | | | Professor, Engineering Science | | | Mirza…
Eric Hermes, Khachik Sargsyan, Habib Najm, Judit Zádor
We present a new algorithm for the optimization of molecular structures to saddle points on the potential energy surface using a redundant internal coordinate system. This algorithm automates the procedure of defining the internal coordinate system, including the handling of linear bending angles, e.g. through the…
Moshik Hershcovitch, Andrew W. Wood, Leshem Choshen, Guy Girmonsky + 6 more
'Roy Leibovitz' 'Ilias Ennmouri' 'Michal Malka' 'Peter Chin' 'Sundararaman Swaminathan' 'Danny Harnik'] With the growth of model sizes and the scale of their deployment, their sheer size burdens the infrastructure requiring more network and more storage to accommodate these. While there is a vast model compression…
Herbert J. Bernstein, Alexei S. Soares, Kimberly Horvat, Jean Jakoncic
New higher-count-rate, integrating, large area X-ray detectors with framing rates as high as 17,400 images per second are beginning to be available. These will soon be used for specialized MX experiments but will require optimal lossy compression algorithms to enable systems to keep up with data throughput. Some…
Michael Hutcheon, Andrew Teale
Algorithms are presented for performing a topological analysis of an arbitrary function, evaluated on an arbitrary grid of points. These algorithms work strictly by post-processing the data and require no additional function evaluations. This is achieved by connecting the grid points with a neighbourhood graph…