Related papers: ABEL: The Adaptable Beginning-to-End Linac simulat…
High-gain free-electron lasers (FELs) are becoming important light sources at short wavelengths such as the EUV and X-ray regimes. A particularly promising concept is the regenerative amplifier FEL (RAFEL), which can greatly increase the…
In recent years, end-to-end autonomous driving architectures have gained increasing attention due to their advantage in avoiding error accumulation. Most existing end-to-end autonomous driving methods are based on Imitation Learning (IL),…
Approximate Bayes Computations (ABC) are used for parameter inference when the likelihood function of the model is expensive to evaluate but relatively cheap to sample from. In particle ABC, an ensemble of particles in the product space of…
GPUs and other accelerators are increasingly used for scientific computing. In the future, we want to add GPU support to parallel adaptive mesh refinement (AMR) codes written in Fortran. To understand which changes are necessary to obtain…
A 52MHz Radio Frequency Quadrupole (RFQ) linear accelerator (linac) is designed to serve as an initial structure for the SSC-linac system (injector into Separated Sector Cyclotron). The designed injection and output energy are 3.5 keV/u and…
Deep learning (DL) models are piquing high interest and scaling at an unprecedented rate. To this end, a handful of tiled accelerators have been proposed to support such large-scale training tasks. However, these accelerators often…
Current physics models used to interpret experimental measurements of particle beams require either simplifying assumptions to be made in order to ensure analytical tractability, or black box optimization methods to perform model based…
This paper presents a heterogeneous adaptive mesh refinement (AMR) framework for efficient simulation of moderately stiff reactive problems. This framework features an elaborate subcycling-in-time algorithm along with a specialized…
Tabular foundation models, such as TabPFNv2 and TabICL, have recently dethroned gradient-boosted trees at the top of predictive benchmarks, demonstrating the value of in-context learning for tabular data. We introduce TabICLv2, a new…
Computational biology has increasingly turned to agent-based modeling to explore complex biological systems. Biological diffusion (diffusion, decay, secretion, and uptake) is a key driver of biological tissues. GPU computing can vastly…
Analog in-memory computing (AIMC) cores offers significant performance and energy benefits for neural network inference with respect to digital logic (e.g., CPUs). AIMCs accelerate matrix-vector multiplications, which dominate these…
The Drive Beam Linac of the Compact Linear Collider (CLIC) has to accelerate an electron beam with 4.2 A up to 2.4 GeV in almost fully-loaded structures. The pulse contains about 70000 bunches, one in every second rf bucket, and has a…
Calibration is a vital step in the development of rigorous digital models of diverse physical and chemical processes, yet one which is highly time- and labour-intensive. In this paper, we introduce a novel tool, Autonomous Calibration and…
An S-band linear accelerator is the source of particles and the front end of the Advanced Photon Source injector. In addition, it supports a low-energy undulator test line (LEUTL) and drives a free-electron laser (FEL). A…
Pedestal modelling is crucial to predict the performance of future fusion devices. Current modelling efforts suffer either from a lack of kinetic physics, or an excess of computational complexity. To ameliorate these problems, we take a…
Particle acceleration in collisionless plasma systems is a central question in astroplasma and astroparticle physics. The structure of the acceleration regions, electron-ion energy equilibration, preacceleration of particles at shocks to…
The simulation of turbulence in the boundary region of a tokamak is crucial for understanding and optimizing the performance of fusion reactors. In this work, the use of low-rank linear algebra techniques is shown to enhance the efficiency…
First-principles atomistic simulations are essential for understanding complex material phenomena but are fundamentally limited by their computational cost. While Machine Learning Interatomic Potentials (MLIPs) have drastically improved…
Pipeline parallelism is widely used to train large language models (LLMs). However, increasing heterogeneity in model architectures exacerbates pipeline bubbles, thereby reducing training efficiency. Existing approaches overlook the…
The demands on performance of advanced linear accelerator based facilities strongly depend on the quality of the particle beams produced by such machines. Indeed, state-of-the-art applications in photon production and high-energy physics…