5 papers
Training Under Challenge: Executable Certificates and Challenge-Closed Optimality for Neural Networks
Farhang Yeganegi, Arian Eamaz, Mojtaba Soltanalian
A flat training curve does not reveal whether a neural network has reached a global optimum, is locally trapped, is representation-limited, or is mismatched to its trainer. We intr…
Trust, but Verify: Peeling Low-Bit Transformer Networks for Training Monitoring
Arian Eamaz, Farhang Yeganegi, Mojtaba Soltanalian
Understanding whether deep neural networks are effectively optimized remains challenging, as training occurs in highly nonconvex landscapes and standard metrics provide limited vis…
From Quasi-Isometric Embeddings to Finite-Volume Property: A Theoretical Framework for Quantized Matrix Completion
Arian Eamaz, Farhang Yeganegi, Mojtaba Soltanalian
We delve into the impact of memoryless scalar quantization on matrix completion. Our primary motivation for this research is to evaluate the recovery performance of nuclear norm mi…
Physics-Inspired Binary Neural Networks: Interpretable Compression with Theoretical Guarantees
Arian Eamaz, Farhang Yeganegi, Mojtaba Soltanalian
Why rely on dense neural networks and then blindly sparsify them when prior knowledge about the problem structure is already available? Many inverse problems admit algorithm-unroll…
Sample Abundance for Signal Processing: A Brief Introduction
Arian Eamaz, Farhang Yeganegi, Mojtaba Soltanalian
This paper reports, by way of introduction, on the advances made by our group and the broader signal processing community on the concept of sample abundance; a phenomenon that natu…