arXiv subjects

Tim Tuuva

Publications and source records attributed to Tim Tuuva.

3 recordsLinked to original sources

arXiv2026

Multi-channel Optical Vision Model

Spatial multiplexing is one of the natural strengths of optics, yet in optical neural networks, it is often used mainly as parallel throughput. Here, we show that spatial multiplexing in an optical neural network can be used not only to process multiple inputs in parallel, but also to define a trainable representational coordinate of the model. In three implemented scenarios, parallel-input processing, class-code readout and channel-mixed feature interaction, spatial channels act as independent learners, structured code dimensions, and interacting feature groups. The programmable free-space optical processor is trained through an online physical-forward/surrogate-backward scheme, where measured optical outputs define the forward pass while a differentiable surrogate estimates gradients and is continually fine-tuned during training from newly acquired optical data. We demonstrate these channel roles in image classification and regression tasks using multi-layer architectures with more than one million trainable optical phase parameters. We further implement a hybrid optical-electronic vision-language model, in which the optical neural network provides visual tokens to a digital transformer decoder for controlled image-captioning tasks. These results establish spatially multiplexed optical channels as a programmable feature and readout space for hybrid optical vision models.

physics.optics↗

arXiv2026

Classical Analog Emulation of Quantum Circuits via Time-Averaged Dynamic States

Classical analog hardware that emulates quantum circuits at the gate level offers a route to benchmarking, prototyping, and teaching quantum algorithms. We introduce wavebits, classical wave analogs of qubits whose amplitudes are carried by physical oscillatory signals, and show that any nonseparable N-qubit state can be encoded in 2N narrowband signals that remain locally separable at every instant. The nonseparable correlations are recovered at readout by time-averaged demodulation over auxiliary carrier frequencies, referred to as nonseparability channels. We prove that any two-qubit gate unravels into the time-averaged tensor product of two local time-varying operators, and that arbitrary circuits are emulated with a base-frequency count scaling linearly with the number of entangling layers, independent of qubit number. The exponential cost of the 2^N-dimensional state reappears at readout and in the averaging time of deep circuits, not during circuit execution, as quantified by an analytic error bound that also serves as a hardware design rule. A mixed-signal prototype emulates Bell state generation, controlled-NOT gates, phase kickback, and Bloch-sphere rotations with fidelities above 0.98, and numerical benchmarks against exact state vectors validate the scheme for up to six qubits. The architecture is directly implementable in acoustic, photonic, and mixed-signal platforms.

quant-ph↗

arXiv2025

Tutorial: A practical guide to the alignment of defocused spatial light modulators for fast diffractive neural networks

The conjugation of multiple spatial light modulators (SLMs) enables the construction of optical diffractive neural networks (DNNs). To accelerate training, which is limited by the low refresh rate of SLMs, spatial multiplexing of the input data across different spatial channels is possible, maximizing the number of available spatial degrees of freedom (DoFs). Precise alignment is required in order to ensure that the same physical operation is performed across each channel and thus the learning operation of the network. We present a semi-automatic procedure for this experimentally challenging alignment resulting in a pixel-level conjugation. It is scalable to any number of SLMs and may be useful in wavefront shaping setups where precise conjugation of SLMs is required, e.g. for the control of optical waves in phase and amplitude. The resulting setup functions as an optical DNN capable of processing hundreds of inputs simultaneously, thereby reducing training times and experimental noise through spatial averaging. We further present a characterization of the setup and an alignment method.

physics.optics↗