Search Results - "Tsakyridis, A."

An all-optical neuron with sigmoid activation function by Mourgias-Alexandris, G, Tsakyridis, A, Passalis, N, Tefas, A, Vyrsokinos, K, Pleros, N

“…We present an all-optical neuron that utilizes a logistic sigmoid activation function, using a Wavelength-Division Multiplexing (WDM) input & weighting scheme…”
Get full text

Noise-resilient and high-speed deep learning with coherent silicon photonics by Mourgias-Alexandris, G., Moralis-Pegios, M., Tsakyridis, A., Simos, S., Dabos, G., Totovic, A., Passalis, N., Kirtas, M., Rutirawut, T., Gardes, F. Y., Tefas, A., Pleros, N.

“…The explosive growth of deep learning applications has triggered a new era in computing hardware, targeting the efficient deployment of multiply-and-accumulate…”
Get full text

Channel response-aware photonic neural network accelerators for high-speed inference through bandwidth-limited optics by Mourgias-Alexandris, G, Moralis-Pegios, M, Tsakyridis, A, Passalis, N, Kirtas, M, Tefas, A, Rutirawut, T, Gardes, F Y, Pleros, N

“…Photonic neural network accelerators (PNNAs) have been lately brought into the spotlight as a new class of custom hardware that can leverage the maturity of…”
Get full text

Performance analysis of a 1024-port Hipoλaos OPS in DCN, HPC, and 5G fronthauling Ethernet applications by Terzenidis, N., Giamougiannis, G., Tsakyridis, A., Spasopoulos, D., Yan, F., Calabretta, N., Vagionas, C., Pleros, N.

“…The explosive traffic growth of emerging cloud, augmented/virtual reality, artificial intelligence, and 5G applications along with the inherent need for…”
Get full text

Fiber Wireless A-RoF/IFoF Uplink of 0.4Gb/s 16-QAM and 0.6Gb/s QPSK Over a 32-Element 60GHz Phased Array Antenna for 5G Fronthaul Networks by Ruggeri, E., Tsakyridis, A., Kalfas, G., Miliou, A., Pleros, N., Vagionas, C., Leiba, Y.

“…We experimentally demonstrate an end-to-end Fiber Wireless A-RoF/IFoF fronthaul link using a 32-element 60GHz phased array antenna and characterize for the…”
Get full text

End-to-end optical packet switching with burst-mode reception at 25 Gb/s through a 1024-port 25.6 Tb/s capacity \text\lambda\text Optical Packet Switch by Tsakyridis, A., Terzenidis, N., Giamougiannis, G., van Kerrebrouck, J., Verbeke, M., Torfs, G., Moralis-Pegios, M., Pleros, N.

“…We demonstrate end-to-end 25Gb/s true optical packet switching featuring burst-mode reception with <50ns locking time through a 1024-port 25.6Tb/s capacity…”
Get full text

Optical Content Addressable Memory Matchline for 2-bit Address Look-Up at 10 Gb/s by Mourgias-Alexandris, G., Vagionas, C., Tsakyridis, A., Maniotis, P., Pleros, N.

“…We demonstrate the first experimental evaluation of a 10Gb/s optical matchline (ML) memory architecture that designates successfully whether an incoming 2-bit…”
Get full text

Programmable Tanh-, ELU-, Sigmoid-, and Sin-based Nonlinear Activation Functions for Neuromorphic Photonics by Pappas, C., Kovaios, S., Moralis-Pegios, M., Tsakyridis, A., Giamougiannis, G., Kirtas, M., Kerrebrouck, J. Van, Coudyzer, G., Yin, X., Passalis, N., Tefas, A., Pleros, N.

“…We demonstrate a programmable analog opto-electronic (OE) circuit that can be configured to provide a range of nonlinear activation functions for incoherent…”
Get full text

A 160 TOPS Multi-dimensional AWGR-based accelerator for Deep Learning by Pappas, C., Moschos, T., Prapas, A., Tsakyridis, A., Moralis-Pegios, M., Vyrsokinos, K., Pleros, N.

“…We experimentally demonstrate a novel time-space-wavelength multiplexed AWGR-based photonic tensor multiplier that provides a total computational power of…”
Get full text

Photonic Neural Networks and Optics-informed Deep Learning Fundamentals by Tsakyridis, A, Moralis-Pegios, M, Giamougiannis, G, Kirtas, M, Passalis, N, Tefas, A, Pleros, N

“…The recent explosive compute growth, mainly fueled by the boost of AI and DNNs, is currently instigating the demand for a novel computing paradigm that can…”
Get full text

Sub-pJ/MAC Silicon Photonic GeMM for Optical Neural Networks using a Time-Space Multiplexed Coherent Xbar by Kovaios, S., Roumpos, I., Tsakyridis, A., Giamougiannis, G., Pegios, M. Moralis, Berciano, M., Ferraro, F., Bode, D., Srinivasan, A., Pantouvaki, M., Pleros, N.

“…We present a time-space multiplexed Silicon Photonic Neural Network that acts as a General Matrix Multiply (GeMM) engine, using a 2×2 photonic Xbar prototype…”
Get full text

Optics-informed Neural Networks: Bridging Deep Learning with Photonic Accelerators by Moralis-Pegios, M., Tsakyridis, A., Pappas, C., Moschos, T., Giamougiannis, G., Kovaios, S., Roumpos, I., Kirtas, M., Passalis, N., Tefas, A., Pleros, N.

“…We discuss our work in optics informed photonic neural networks, an architectural framework bridging the idiosyncrasy of integrated photonic architectures with…”
Get full text

Silicon Integrated Photonic-Electronic Multiply-Accumulate Neurons by Roumpos, I., De Marinis, L., Kincaid, P. S., Paolini, E., Kovaios, S., Tsakyridis, A., Moralis-Pegios, M., Berciano, M., Ferraro, F., Bode, D., Srinivasan, A., Pantouvaki, M., Contestabile, G., Vyrsokinos, K., Pleros, N.

“…We experimentally demonstrate an EAM-based photonic-electronic multiply-accumulate neuron that performs inference in a health monitoring task with 1350…”
Get full text

A 10Gb/S Optical Random-Access Memory Using a Saturated SOA-MZI Fast Access Gate and a Monolithic InP Flip-Flop by Tsakyridis, A., Vagionas, C., Alexoudi, Th, Miliou, A., Pleros, N.

“…We experimentally demonstrate an all-optical static Random-Access Memory (RAM)cell using a monolithic integrated InP Flip-Flop and a fast strongly-saturated…”
Get full text

A 25.6 Tbps capacity 1024-port Hipoλaos Optical Packet Switch Architecture for disaggregated datacenters by Terzenidis, N., Tsakyridis, A., Giamougiannis, G., Moralis-Pegios, M., Vyrsokinos, K., Pleros, N.

“…We demonstrate experimentally the feasibility of a 25.6Tb/s capacity Hipoλaos optical packet switch architecture with 1024 in/out ports operating at 25Gb/s,…”
Get full text

A 10Gb/s All-Optical Match-line for optical Content Addressable Memory (CAM) Rows by Mourgias-Alexandris, G., Vagionas, C., Tsakyridis, A., Maniotis, P., Pleros, N.

“…We experimentally demonstrate the first all-optical match-line for a two Optical CAM-cell-based address look-up row, using wavelength encoding and an…”
Get full text

A Flexible and Reconfigurable Si3N4 ROADM-enabled 5G mmWave IFoF Fiber Wireless Fronthaul with 60 GHz beamsteering capabilities by Tsakyridis, A., Ruggeri, E., Kalfas, G., Oldenbeuving, R.M., van Dijk, P.W.L., Roeloffzen, C., Leiba, Y., Miliou, A., Pleros, N., Vagionas, C.

“…A bandwidth-reconfigurable mmWave Fiber Wireless fronthaul bus topology of four 1 Gb/s 16-QAM streams transmitted through two cascaded low-loss Si 3 N 4 -ROADM…”
Get full text

All-optical Ternary Content Addressable Memory (T-CAM) Cell for ultra-fast Address Look-ups in Router Applications by Mourgias-Alexandris, G., Vagionas, C., Tsakyridis, A., Maniotis, P., Pleros, N.

“…We experimentally demonstrate the first all-optical Ternary Content Addressable Memory (T-CAM) cell based on two monolithic integrated InP Flip-Flops and a…”
Get full text

Lossless Silicon Photonic ROADM based on a Si3N4 platform and a Monolithically Integrated Erbium Doped Amplifier by Vagionas, C., Tsakyridis, A., Chrysostomidis, T., Roumpos, I., Fotiadis, K., Manolis, A., Mu, J., Dijkstra, M., Blanco, SM Garcia, Oldenbeuving, R. M., van Dijk, P. W. L., Roeloffzen, C. G. H., Vyrsokinos, K., Pleros, N., Alexoudi, T.

“…The first demonstration of a lossless four-port silicon photonic ROADM-node based on a monolithic-integrated spiral Al 2 O 3 :Er 3+ Erbium Doped Waveguide…”
Get full text

A Linear Photonic Neuron Using an Extended Iq Modulator Cell for Enabling All-Optical Coherent Sigmoid Neurons by Mourgias-Alexandris, G, Totovic, A, Tsakyridis, A, Passalis, N, Tefas, A, Vyrsokinos, K, Pleros, N

“…We present a single-wavelength coherent linear neuron for allowing both weight amplitude and phase-encoded sign representation. All-optical algebraic summation…”
Get full text