Ellen Zhou

Neuromorphic photonic networks using silicon photonic weight banks

Photonic systems for high-performance information processing have attracted renewed interest. Neuromorphic silicon photonics has the potential to integrate processing functions that vastly exceed the capabilities of electronics. We report first observations of a recurrent silicon photonic neural network, in which connections are configured by microring weight banks. A mathematical isomorphism between the silicon photonic circuit and a continuous neural network model is demonstrated through dynamical bifurcation analysis. Exploiting this isomorphism, a simulated 24-node silicon photonic neural network is programmed using “neural compiler” to solve a differential system emulation task. A 294-fold acceleration against a conventional benchmark is predicted. We also propose and derive power consumption analysis for modulator-class neurons that, as opposed to laser-class neurons, are compatible with silicon photonic platforms. At increased scale, Neuromorphic silicon photonics could access new regimes of ultrafast information processing for radio, control, and scientific computing.Photonic systems for high-performance information processing have attracted renewed interest. Neuromorphic silicon photonics has the potential to integrate processing functions that vastly exceed the capabilities of electronics. We report first observations of a recurrent silicon photonic neural network, in which connections are configured by microring weight banks. A mathematical isomorphism between the silicon photonic circuit and a continuous neural network model is demonstrated through dynamical bifurcation analysis. Exploiting this isomorphism, a simulated 24-node silicon photonic neural network is programmed using “neural compiler” to solve a differential system emulation task. A 294-fold acceleration against a conventional benchmark is predicted. We also propose and derive power consumption analysis for modulator-class neurons that, as opposed to laser-class neurons, are compatible with silicon photonic platforms. At increased scale, Neuromorphic silicon photonics could access new regimes of ultrafast information processing for radio, control, and scientific computing.

Microring Weight Banks

Microring weight banks could enable novel signal processing approaches in silicon photonics. We analyze factors limiting channel count in microring weight banks, which are central to analog wavelength-division multiplexed processing networks in silicon. We find that microring weight banks require a fundamentally different analysis compared to other wavelength-division multiplexing circuits (e.g., demultiplexers). By introducing a quantitative description of independent weighting, we establish performance tradeoffs between channel count and power penalty. This performance is significantly affected by coherent multiresonator interactions through bus waveguides. We experimentally demonstrate these effects in a fabricated device. Analysis relies on the development of a novel simulation technique combining parametric programming with generalized transmission theory. Experimental measurement fitting of an 8-channel weight bank is presented as an example of another application of the simulator.

Balanced WDM weight banks for analog optical processing and networking in silicon

We demonstrate complementary (+/-) WDM weighted addition in a bank of silicon microrings using a balanced detection technique. Weighted addition that is tunable and complementary is a key function for multivariate analog signal processing and could enable scalable analog networking approaches with silicon photonic technologies.

Coherent interactions in microring weight banks and impact on channel density

We experimentally observe and simulate coherent inter-resonator effects specific to microring weight banks. An analysis based on this effect results in quantitative performance limits of weight banks, a key subcircuit for multivariate analog signal processing and scalable analog interconnect approaches in silicon photonics.

Multi-channel microring weight bank control for reconfigurable analog photonic networks

We demonstrate 4-channel weighted addition in a silicon microring filter bank with 3.8 bit accuracy. Scalable calibration techniques are developed for thermal cross-talk and cross-gain saturation. Practical weight control is essential for large-scale photonic processing based on microrings.

Silicon photonic weight bank control of integrated analog network dynamics

Analog Interconnection networks are configured setting connection weights. Microring weight banks are a key device for making such networks in silicon photonic circuits. We demonstrate a small analog network in silicon using a single node with simple dynamics, showing dynamics parameterized by microring weight banks.

Demonstration of a silicon photonic neural network

Silicon photonic integration could enable high-performance brain-inspired photonic processors. We demonstrate a 3 node recurrent photonic neural network. Cusp and Hopf bifurcations induced by synaptic reconfiguration are shown as proof-of-concept. The prototype represents an early step towards network-based models of physical computing with integrated photonics.