FLOPS: EFficient On-Chip Learning for OPtical Neural Networks Through Stochastic Zeroth-Order Optimization

Abstract Deeplearning algorithms are revolutionising many aspects of modern life. Typically, they are implemented in CMOS-based hardware with severely limited memory access times and inefficient data-routing. All-optical neural networks without any electro-optic conversions could alleviate these shortcomings. However, an all-optical nonlinear activation function, which is a vital building block for optical neural networks, needs to be developed efficiently on-chip. Here, we introduce and demonstrate both optical synapse weighting and all-optical nonlinear thresholding using two different effects in one single chalcogenide material. We show how the structural phase transitions in a wide-bandgap phase-change material enables storing the neural network weights via non-volatile photonic memory, whilst resonant bond destabilisation is used as a nonlinear activation threshold without changing the material. These two different transitions within chalcogenides enable programmable neural networks with near-zero static power consumption once trained, in addition to picosecond delays performing inference tasks not limited by wire charging that limit electrical circuits; for instance, we show that nanosecond-order weight programming and near-instantaneous weight updates enable accurate inference tasks within 20 picoseconds in a 3-layer all-optical neural network. Optical neural networks that bypass electro-optic conversion altogether hold promise for network-edge machine learning applications where decision-making in real-time are critical, such as for autonomous vehicles or navigation systems such as signal pre-processing of LIDAR systems.

Download Full-text

Predicting Networks-on-Chip traffic congestion with Spiking Neural Networks

Journal of Parallel and Distributed Computing ◽

10.1016/j.jpdc.2021.03.013 ◽

2021 ◽

Author(s):

Aqib Javed ◽

Jim Harkin ◽

Liam McDaid ◽

Junxiu Liu

Keyword(s):

Neural Networks ◽

Traffic Congestion ◽

Spiking Neural Networks ◽

Networks On Chip ◽

On Chip

Download Full-text

A Scalable System-on-Chip Acceleration for Deep Neural Networks

IEEE Access ◽

10.1109/access.2021.3094675 ◽

2021 ◽

pp. 1-1

Author(s):

Faisal Shehzad ◽

Muhammad Rashid ◽

Mohammed H Sinky ◽

Saud S Alotaibi ◽

Muhammad Yousuf Irfan Zia

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

System On Chip ◽

Scalable System ◽

On Chip

Download Full-text

Optical Axons for Electro-Optical Neural Networks

Sensors ◽

10.3390/s20216119 ◽

2020 ◽

Vol 20 (21) ◽

pp. 6119

Author(s):

Mircea Hulea ◽

Zabih Ghassemlooy ◽

Sujan Rajbhandari ◽

Othman Isam Younus ◽

Alexandru Barleanu

Keyword(s):

Neural Networks ◽

Optical Signal ◽

Sensor Data ◽

Line Of Sight ◽

Optical Intensity ◽

Digital Signals ◽

Link Length ◽

Optical Neural Networks ◽

Analogue Signals ◽

Neural Unit

Recently, neuromorphic sensors, which convert analogue signals to spiking frequencies, have been reported for neurorobotics. In bio-inspired systems these sensors are connected to the main neural unit to perform post-processing of the sensor data. The performance of spiking neural networks has been improved using optical synapses, which offer parallel communications between the distanced neural areas but are sensitive to the intensity variations of the optical signal. For systems with several neuromorphic sensors, which are connected optically to the main unit, the use of optical synapses is not an advantage. To address this, in this paper we propose and experimentally verify optical axons with synapses activated optically using digital signals. The synaptic weights are encoded by the energy of the stimuli, which are then optically transmitted independently. We show that the optical intensity fluctuations and link’s misalignment result in delay in activation of the synapses. For the proposed optical axon, we have demonstrated line of sight transmission over a maximum link length of 190 cm with a delay of 8 μs. Furthermore, we show the axon delay as a function of the illuminance using a fitted model for which the root mean square error (RMS) similarity is 0.95.

Download Full-text