Related papers: ELT-scale Adaptive Optics real-time control with t…
Purpose: To develop new encoding and reconstruction techniques for fast multi-contrast quantitative imaging. Methods: The recently proposed Echo Planar Time-resolved Imaging (EPTI) technique can achieve fast distortion- and blurring-free…
Emerging AI applications such as ChatGPT, graph convolutional networks, and other deep neural networks require massive computational resources for training and inference. Contemporary computing platforms such as CPUs, GPUs, and TPUs are…
The demand for edge AI in vision-language tasks requires models that achieve real-time performance on resource-constrained devices with limited power and memory. This paper proposes two adaptive compression techniques -- Sparse Temporal…
We investigate and characterize the performance of an important class of operations on GPUs and Many Integrated Core (MIC) architectures. Our work is motivated by applications that analyze low-dimensional spatial datasets captured by high…
Discord is a refinement of the concept of anomalous subsequence of a time series. The task of discords discovery is applied in a wide range of subject domains related to time series: medicine, economics, climate modeling, etc. In this…
Spiking Neural Networks (SNNs) often suffer from high time complexity $O(T)$ due to the sequential processing of $T$ spikes, making training computationally expensive. In this paper, we propose a novel Fixed-point Parallel Training (FPT)…
Vision Transformers (ViTs) have emerged as a powerful architecture for computer vision tasks due to their ability to model long-range dependencies and global contextual relationships. However, their substantial compute and memory demands…
Large model inference is shifting from cloud to edge due to concerns about the privacy of user interaction data. However, edge devices often struggle with limited computing power, memory, and bandwidth, requiring collaboration across…
Convolution layers are prevalent in many classes of deep neural networks, including Convolutional Neural Networks (CNNs) which provide state-of-the-art results for tasks like image recognition, neural machine translation and speech…
Neural networks commonly execute on hardware accelerators such as NPUs and GPUs for their size and computation overhead. These accelerators are costly and it is hard to scale their resources to handle real-time workload fluctuations. We…
Atmospheric wavefront prediction based on previous wavefront sensor measurements can greatly enhance the performance of adaptive optics systems. We propose an optimal linear approach based on the Empirical Orthogonal Functions (EOF)…
HARMONI is the first light visible and near-IR integral field spectrograph for the ELT covering a large spectral range from 450nm to 2450nm with resolving powers from 3500 to 18000 and spatial sampling from 60mas to 4mas. It can operate in…
Modern UAV architectures increasingly aim to unify high-level autonomy and low-level flight control on a single General-Purpose Operating System (GPOS). However, complex multi-core System-on-Chips (SoCs) introduce significant timing…
Heterogeneous parallel error detection is an approach to achieving fault-tolerant processors, leveraging multiple power-efficient cores to re-execute software originally run on a high-performance core. Yet, its complex components, gathering…
We present the results obtained with an end-to-end simulator of an Extreme Adaptive Optics (XAO) system control loop. It is used to predict its on-sky performances and to optimise the AO loop algorithms. It was first used to validate a…
Ant Colony algorithm has been applied to various optimization problems, however most of the previous work on scaling and parallelism focuses on Travelling Salesman Problems (TSPs). Although, useful for benchmarks and new idea comparison,…
The Efficient Adaptive Transformer (EAT) framework unifies three adaptive efficiency techniques - progressive token pruning, sparse attention, and dynamic early exiting - into a single, reproducible architecture for input-adaptive…
The next generation of Extremely Large Telescopes (ELTs), with their wide apertures and advanced Multi-Conjugate Adaptive Optics (MCAO) systems, will provide unprecedented sharp and deep observations, even surpassing the capabilities of…
Co-expression network is a critical technique for the identification of inter-gene interactions, which usually relies on all-pairs correlation (or similar measure) computation between gene expression profiles across multiple samples.…
Parameter-efficient transfer learning (PETL) has shown great potential in adapting a vision transformer (ViT) pre-trained on large-scale datasets to various downstream tasks. Existing studies primarily focus on minimizing the number of…