Showing 1–4 of 4 results for author: Jost, D

A 262 TOPS Hyperdimensional Photonic AI Accelerator powered by a Si3N4 microcomb laser

Authors: Christos Pappas, Antonios Prapas, Theodoros Moschos, Manos Kirtas, Odysseas Asimopoulos, Apostolos Tsakyridis, Miltiadis Moralis-Pegios, Chris Vagionas, Nikolaos Passalis, Cagri Ozdilek, Timofey Shpakovsky, Alain Yuji Takabayashi, John D. Jost, Maxim Karpov, Anastasios Tefas, Nikos Pleros

Abstract: The ever-increasing volume of data has necessitated a new computing paradigm, embodied through Artificial Intelligence (AI) and Large Language Models (LLMs). Digital electronic AI computing systems, however, are gradually reaching their physical plateaus, stimulating extensive research towards next-generation AI accelerators. Photonic Neural Networks (PNNs), with their unique ability to capitalize… ▽ More The ever-increasing volume of data has necessitated a new computing paradigm, embodied through Artificial Intelligence (AI) and Large Language Models (LLMs). Digital electronic AI computing systems, however, are gradually reaching their physical plateaus, stimulating extensive research towards next-generation AI accelerators. Photonic Neural Networks (PNNs), with their unique ability to capitalize on the interplay of multiple physical dimensions including time, wavelength, and space, have been brought forward with a credible promise for boosting computational power and energy efficiency in AI processors. In this article, we experimentally demonstrate a novel multidimensional arrayed waveguide grating router (AWGR)-based photonic AI accelerator that can execute tensor multiplications at a record-high total computational power of 262 TOPS, offering a ~24x improvement over the existing waveguide-based optical accelerators. It consists of a 16x16 AWGR that exploits the time-, wavelength- and space- division multiplexing (T-WSDM) for weight and input encoding together with an integrated Si3N4-based frequency comb for multi-wavelength generation. The photonic AI accelerator has been experimentally validated in both Fully-Connected (FC) and Convolutional NN (NNs) models, with the FC and CNN being trained for DDoS attack identification and MNIST classification, respectively. The experimental inference at 32 Gbaud achieved a Cohen's kappa score of 0.867 for DDoS detection and an accuracy of 92.14% for MNIST classification, respectively, closely matching the software performance. △ Less

Submitted 5 March, 2025; originally announced March 2025.

Univariate Radial Basis Function Layers: Brain-inspired Deep Neural Layers for Low-Dimensional Inputs

Authors: Daniel Jost, Basavasagar Patil, Xavier Alameda-Pineda, Chris Reinke

Abstract: Deep Neural Networks (DNNs) became the standard tool for function approximation with most of the introduced architectures being developed for high-dimensional input data. However, many real-world problems have low-dimensional inputs for which standard Multi-Layer Perceptrons (MLPs) are the default choice. An investigation into specialized architectures is missing. We propose a novel DNN layer call… ▽ More Deep Neural Networks (DNNs) became the standard tool for function approximation with most of the introduced architectures being developed for high-dimensional input data. However, many real-world problems have low-dimensional inputs for which standard Multi-Layer Perceptrons (MLPs) are the default choice. An investigation into specialized architectures is missing. We propose a novel DNN layer called Univariate Radial Basis Function (U-RBF) layer as an alternative. Similar to sensory neurons in the brain, the U-RBF layer processes each individual input dimension with a population of neurons whose activations depend on different preferred input values. We verify its effectiveness compared to MLPs in low-dimensional function regressions and reinforcement learning tasks. The results show that the U-RBF is especially advantageous when the target function becomes complex and difficult to approximate. △ Less

Submitted 2 February, 2024; v1 submitted 7 November, 2023; originally announced November 2023.

Learning battery model parameter dynamics from data with recursive Gaussian process regression

Authors: Antti Aitio, Dominik Jöst, Dirk Uwe Sauer, David A. Howey

Abstract: Estimating state of health is a critical function of a battery management system but remains challenging due to the variability of operating conditions and usage requirements of real applications. As a result, techniques based on fitting equivalent circuit models may exhibit inaccuracy at extremes of performance and over long-term ageing, or instability of parameter estimates. Pure data-driven tec… ▽ More Estimating state of health is a critical function of a battery management system but remains challenging due to the variability of operating conditions and usage requirements of real applications. As a result, techniques based on fitting equivalent circuit models may exhibit inaccuracy at extremes of performance and over long-term ageing, or instability of parameter estimates. Pure data-driven techniques, on the other hand, suffer from lack of generality beyond their training dataset. In this paper, we propose a hybrid approach combining data- and model-driven techniques for battery health estimation. Specifically, we demonstrate a Bayesian data-driven method, Gaussian process regression, to estimate model parameters as functions of states, operating conditions, and lifetime. Computational efficiency is ensured through a recursive approach yielding a unified joint state-parameter estimator that learns parameter dynamics from data and is robust to gaps and varying operating conditions. Results show the efficacy of the method, on both simulated and measured data, including accurate estimates and forecasts of battery capacity and internal resistance. This opens up new opportunities to understand battery ageing in real applications. △ Less

Submitted 26 April, 2023; originally announced April 2023.

Application of reactive transport modelling to growth and transport of microorganisms in the capillary fringe

Authors: Pavel Hron, Daniel Jost, Peter Bastian, Claudia Gallert, Josef Winter, Olaf Ippisch

Abstract: A multicomponent multiphase reactive transport simulator has been developed to facilitate the investigation of a large variety of phenomena in porous media including component transport, diffusion, microbiological growth and decay, cell attachment and detachment and phase exchange. The coupled problem is solved using operator splitting. This approach allows a flexible adaptation of the solution st… ▽ More A multicomponent multiphase reactive transport simulator has been developed to facilitate the investigation of a large variety of phenomena in porous media including component transport, diffusion, microbiological growth and decay, cell attachment and detachment and phase exchange. The coupled problem is solved using operator splitting. This approach allows a flexible adaptation of the solution strategy to the concrete problem. Moreover, the individual submodels were optimised to be able to describe behaviour of Escherichia coli (HB101 K12 pGLO) in the capillary fringe in the presence or absence of dissolved organic carbon and oxygen under steady-state and flow conditions. Steady-state and flow through experiments in a Hele-Shaw cell, filled with quartz sand, were conducted to study eutrophic bacterial growth and transport in both saturated and unsaturated porous media. As E. coli cells can form the green fluorescent protein (GFP), the cell densities, calculated by evaluation of measured fluorescence intensities (in situ detection) were compared with the cell densities computed by numerical simulation. The comparison showed the laboratory experiments can be well described by our mathematical model. △ Less

Submitted 23 October, 2014; originally announced October 2014.