187 citations · 221 across the 12 of their papers we have counts for
12 papers
Exploring Expert Specialization through Unsupervised Training in Sparse Mixture of Experts
Strahinja Nikolic, Ilker Oguz, Demetri Psaltis
Understanding the internal organization of neural networks remains a fundamental challenge in deep learning interpretability. We address this challenge by exploring a novel Sparse…
Roadmap on Neuromorphic Photonics
Daniel Brunner, Bhavin J. Shastri, Mohammed A. Al Qadasi +147
This roadmap consolidates recent advances while exploring emerging applications, reflecting the remarkable diversity of hardware platforms, neuromorphic concepts, and implementatio…
Training Hybrid Neural Networks with Multimode Optical Nonlinearities Using Digital Twins
Ilker Oguz, Louis J. E. Suter, Jih-Liang Hsieh +4
The ability to train ever-larger neural networks brings artificial intelligence to the forefront of scientific and technical discoveries. However, their exponentially increasing si…
Optical Diffusion Models for Image Generation
Ilker Oguz, Niyazi Ulas Dinc, Mustafa Yildirim +6
Diffusion models generate new samples by progressively decreasing the noise from the initially provided random distribution. This inference procedure generally utilizes a trained n…
Training of Physical Neural Networks
Ali Momeni, Babak Rahmani, Benjamin Scellier +25
Physical neural networks (PNNs) are a class of neural-like networks that leverage the properties of physical systems to perform computation. While PNNs are so far a niche research…
Programming the scalable optical learning operator with spatial-spectral optimization
Yi Zhou, Jih-Liang Hsieh, Ilker Oguz +6
Electronic computers have evolved drastically over the past years with an ever-growing demand for improved performance. However, the transfer of information from memory and high en…
Multicasting Optical Reconfigurable Switch
Niyazi Ulas Dinc, Mustafa Yildirim, Ilker Oguz +2
Artificial Intelligence (AI) demands large data flows within datacenters, heavily relying on multicasting data transfers. As AI models scale, the requirement for high-bandwidth and…
Nonlinear Processing with Linear Optics
Mustafa Yildirim, Niyazi Ulas Dinc, Ilker Oguz +2
Deep neural networks have achieved remarkable breakthroughs by leveraging multiple layers of data processing to extract hidden representations, albeit at the cost of large electron…
Forward-Forward Training of an Optical Neural Network
Ilker Oguz, Junjie Ke, Qifei Wang +6
Neural networks (NN) have demonstrated remarkable capabilities in various tasks, but their computation-intensive nature demands faster and more energy-efficient hardware implementa…
Nonlinear Optical Data Transformer for Machine Learning
Mustafa Yildirim, Ilker Oguz, Fabian Kaufmann +4
Modern machine learning models use an ever-increasing number of parameters to train (175 billion parameters for GPT-3) with large datasets to obtain better performance. Bigger is b…
Programming Nonlinear Propagation for Efficient Optical Learning Machines
Ilker Oguz, Jih-Liang Hsieh, Niyazi Ulas Dinc +5
The ever-increasing demand for processing data with larger machine learning models requires more efficient hardware solutions due to limitations such as power dissipation and scala…
Scalable Optical Learning Operator
Uğur Teğin, Mustafa Yıldırım, İlker Oğuz +2
Today's heavy machine learning tasks are fueled by large datasets. Computing is performed with power hungry processors whose performance is ultimately limited by the data transfer…