no code implementations • 13 May 2024 • Kevin Stangl, Marius Arvinte, Weilin Xu, Cory Cornelius
Zero-shot anomaly segmentation using pre-trained foundation models is a promising approach that enables effective algorithms without expensive, domain-specific training or fine-tuning.
no code implementations • 6 May 2024 • Ravikumar Balakrishnan, Marius Arvinte, Nageen Himayat, Hosein Nikopour, Hassnaa Moustafa
A comprehensive modeling of the security threats and the demonstration of adversarial attacks and defenses on practical AI based O-RAN systems is still in its nascent stages.
no code implementations • 10 Oct 2023 • Marius Arvinte, Cory Cornelius, Jason Martin, Nageen Himayat
Beyond their impressive sampling capabilities, score-based diffusion models offer a powerful analysis tool in the form of unbiased density estimation of a query sample under the training data distribution.
no code implementations • 2 May 2023 • Asad Aali, Marius Arvinte, Sidharth Kumar, Jonathan I. Tamir
We present SURE-Score: an approach for learning score-based generative models using training samples corrupted by additive Gaussian noise.
no code implementations • 21 Feb 2023 • Madhumitha Sakthi, Ahmed Tewfik, Marius Arvinte, Haris Vikalo
Automotive radar has increasingly attracted attention due to growing interest in autonomous driving technologies.
2 code implementations • 14 Apr 2022 • Marius Arvinte, Jonathan I Tamir
We introduce a framework for training score-based generative models for wireless MIMO channels and performing channel estimation based on posterior sampling at test time.
no code implementations • 8 Mar 2022 • Madhumitha Sakthi, Ahmed Tewfik, Marius Arvinte, Haris Vikalo
We show robust detection based on radar data reconstructed using 20% of samples under extreme weather conditions such as snow or fog, and on low-illuminated nights.
1 code implementation • 16 Nov 2021 • Marius Arvinte, Jonathan I Tamir
We train a score-based model on channel realizations from the CDL-D model for two antenna spacings and show that the approach leads to competitive in- and out-of-distribution performance when compared to generative adversarial network (GAN) and compressed sensing (CS) methods.
no code implementations • NeurIPS Workshop Deep_Invers 2021 • Ajil Jalal, Marius Arvinte, Giannis Daras, Eric Price, Alex Dimakis, Jonathan Tamir
The CSGM framework (Bora-Jalal-Price-Dimakis'17) has shown that deep generative priors can be powerful tools for solving inverse problems.
1 code implementation • 18 Oct 2021 • Marius Arvinte, Jonathan I. Tamir
Deep learning has been recently applied to physical layer processing in digital communication systems in order to improve end-to-end performance.
2 code implementations • NeurIPS 2021 • Ajil Jalal, Marius Arvinte, Giannis Daras, Eric Price, Alexandros G. Dimakis, Jonathan I. Tamir
The CSGM framework (Bora-Jalal-Price-Dimakis'17) has shown that deep generative priors can be powerful tools for solving inverse problems.
1 code implementation • 2 Mar 2021 • Marius Arvinte, Sriram Vishwanath, Ahmed H. Tewfik, Jonathan I. Tamir
Accelerated multi-coil magnetic resonance imaging reconstruction has seen a substantial recent improvement combining compressed sensing with deep learning.
no code implementations • 23 Dec 2020 • Marius Arvinte, Ahmed H. Tewfik, Sriram Vishwanath
In this work, we introduce EQ-Net: the first holistic framework that solves both the tasks of log-likelihood ratio (LLR) estimation and quantization using a data-driven method.
1 code implementation • 5 Jun 2020 • Marius Arvinte, Ahmed H. Tewfik, Sriram Vishwanath
Our architecture uses a contrastive loss termand a disentangled generative model to sample negative pairs.
1 code implementation • 28 Feb 2020 • Marius Arvinte, Ahmed Tewfik, Sriram Vishwanath
We introduce an adversarial sample detection algorithm based on image residuals, specifically designed to guard against patch-based attacks.
1 code implementation • 18 Jun 2019 • Marius Arvinte, Sriram Vishwanath, Ahmed H. Tewfik
In this work, a deep learning-based quantization scheme for log-likelihood ratio (L-value) storage is introduced.
1 code implementation • 11 Mar 2019 • Marius Arvinte, Ahmed H. Tewfik, Sriram Vishwanath
In this work, a deep learning-based method for log-likelihood ratio (LLR) lossy compression and quantization is proposed, with emphasis on a single-input single-output uncorrelated fading communication setting.