arXiv subjects

Tobi Delbruck

Publications and source records attributed to Tobi Delbruck.

At least 19 recordsLinked to original sources

arXiv2026

NeuroAI and Beyond: Bridging Between Advances in Neuroscience and ArtificialIntelligence

Neuroscience and Artificial Intelligence (AI) have made impressive progress in recent years but remain only loosely interconnected. Based on a workshop convened by the National Science Foundation in August 2025, we identify three fundamental capability gaps in current AI: the inability to interact with the physical world, inadequate learning that produces brittle systems, and unsustainable energy and data inefficiency. We describe the neuroscience principles that address each: co-design of body and controller, prediction through interaction, multi-scale learning with neuromodulatory control, hierarchical distributed architectures, and sparse event-driven computation. We present a research roadmap organized around these principles at near, mid, and long-term horizons. We argue that realizing this program requires a new generation of researchers trained across the boundary between neuroscience and engineering, and describe the institutional conditions: interdisciplinary training, hardware access, community standards, and ethics, needed to support them. We conclude that NeuroAI, neuroscience-informed artificial intelligence, has the potential to overcome limitations of current AI while deepening our understanding of biological neural computation.

q-bio.NC↗

arXiv2025

Towards a physically realistic computationally efficient DVS pixel model

Dynamic Vision Sensor (DVS) event camera models are important tools for predicting camera response, optimizing biases, and generating realistic simulated datasets. Existing DVS models have been useful, but have not demonstrated high realism for challenging HDR scenes combined with adequate computational efficiency for array-level scene simulation. This paper reports progress towards a physically realistic and computationally efficient DVS model based on large-signal differential equations derived from circuit analysis, with parameters fitted from pixel measurements and circuit simulation. These are combined with an efficient stochastic event generation mechanism based on first-passage-time theory, allowing accurate noise generation with timesteps greater than 1000x longer than previous methods

eess.IV↗

arXiv2025

Modulating State Space Model with SlowFast Framework for Compute-Efficient Ultra Low-Latency Speech Enhancement

Deep learning-based speech enhancement (SE) methods often face significant computational challenges when needing to meet low-latency requirements because of the increased number of frames to be processed. This paper introduces the SlowFast framework which aims to reduce computation costs specifically when low-latency enhancement is needed. The framework consists of a slow branch that analyzes the acoustic environment at a low frame rate, and a fast branch that performs SE in the time domain at the needed higher frame rate to match the required latency. Specifically, the fast branch employs a state space model where its state transition process is dynamically modulated by the slow branch. Experiments on a SE task with a 2 ms algorithmic latency requirement using the Voice Bank + Demand dataset show that our approach reduces computation cost by 70% compared to a baseline single-branch network with equivalent parameters, without compromising enhancement performance. Furthermore, by leveraging the SlowFast framework, we implemented a network that achieves an algorithmic latency of just 62.5 μs (one sample point at 16 kHz sample rate) with a computation cost of 100 M MACs/s, while scoring a PESQ-NB of 3.12 and SISNR of 16.62.

eess.AS↗

arXiv2024

DeltaKWS: A 65nm 36nJ/Decision Bio-inspired Temporal-Sparsity-Aware Digital Keyword Spotting IC with 0.6V Near-Threshold SRAM

This paper introduces DeltaKWS, to the best of our knowledge, the first $Δ$RNN-enabled fine-grained temporal sparsity-aware KWS IC for voice-controlled devices. The 65 nm prototype chip features a number of techniques to enhance performance, area, and power efficiencies, specifically: 1) a bio-inspired delta-gated recurrent neural network ($Δ$RNN) classifier leveraging temporal similarities between neighboring feature vectors extracted from input frames and network hidden states, eliminating unnecessary operations and memory accesses; 2) an IIR BPF-based FEx that leverages mixed-precision quantization, low-cost computing structure and channel selection; 3) a 24 kB 0.6 V near-$V_\text{TH}$ weight SRAM that achieves 6.6X lower read power than the foundry-provided SRAM. From chip measurement results, we show that the DeltaKWS achieves an 11/12-class GSCD accuracy of 90.5%/89.5% respectively and energy consumption of 36 nJ/decision in 65 nm CMOS process. At 87% temporal sparsity, computing latency and energy/inference are reduced by 2.4X/3.4X, respectively. The IIR BPF-based FEx, $Δ$RNN accelerator, and 24 kB near-$V_\text{TH}$ SRAM blocks occupy 0.084 mm$^{2}$, 0.319 mm$^{2}$, and 0.381 mm$^{2}$ respectively (0.78 mm$^{2}$ in total).

cs.AR↗

arXiv2024

Steering Prediction via a Multi-Sensor System for Autonomous Racing

Autonomous racing has rapidly gained research attention. Traditionally, racing cars rely on 2D LiDAR as their primary visual system. In this work, we explore the integration of an event camera with the existing system to provide enhanced temporal information. Our goal is to fuse the 2D LiDAR data with event data in an end-to-end learning framework for steering prediction, which is crucial for autonomous racing. To the best of our knowledge, this is the first study addressing this challenging research topic. We start by creating a multisensor dataset specifically for steering prediction. Using this dataset, we establish a benchmark by evaluating various SOTA fusion methods. Our observations reveal that existing methods often incur substantial computational costs. To address this, we apply low-rank techniques to propose a novel, efficient, and effective fusion design. We introduce a new fusion learning policy to guide the fusion process, enhancing robustness against misalignment. Our fusion architecture provides better steering prediction than LiDAR alone, significantly reducing the RMSE from 7.72 to 1.28. Compared to the second-best fusion method, our work represents only 11% of the learnable parameters while achieving better accuracy. The source code, dataset, and benchmark will be released to promote future research.

cs.CV↗

arXiv2024

Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement

This paper introduces a new Dynamic Gated Recurrent Neural Network (DG-RNN) for compute-efficient speech enhancement models running on resource-constrained hardware platforms. It leverages the slow evolution characteristic of RNN hidden states over steps, and updates only a selected set of neurons at each step by adding a newly proposed select gate to the RNN model. This select gate allows the computation cost of the conventional RNN to be reduced during network inference. As a realization of the DG-RNN, we further propose the Dynamic Gated Recurrent Unit (D-GRU) which does not require additional parameters. Test results obtained from several state-of-the-art compute-efficient RNN-based speech enhancement architectures using the DNS challenge dataset, show that the D-GRU based model variants maintain similar speech intelligibility and quality metrics comparable to the baseline GRU based models even with an average 50% reduction in GRU computes.

eess.AS↗

arXiv2024

SciDVS: A Scientific Event Camera with 1.7% Temporal Contrast Sensitivity at 0.7 lux

This paper reports a Dynamic Vision Sensor (DVS) event camera that is 6x more sensitive at 14x lower illumination than existing commercial and prototype cameras. Event cameras output a sparse stream of brightness change events. Their high dynamic range (HDR), quick response, and high temporal resolution provide key advantages for scientific applications that involve low lighting conditions and sparse visual events. However, current DVS are hindered by low sensitivity, resulting from shot noise and pixel-to-pixel mismatch. Commercial DVS have a minimum brightness change threshold of >10%. Sensitive prototypes achieved as low as 1%, but required kilo-lux illumination. Our SciDVS prototype fabricated in a 180nm CMOS image sensor process achieves 1.7% sensitivity at chip illumination of 0.7 lx and 18 Hz bandwidth. Novel features of SciDVS are (1) an auto-centering in-pixel preamplifier providing intrascene HDR and increased sensitivity, (2) improved control of bandwidth to limit shot noise, and (3) optional pixel binning, allowing the user to trade spatial resolution for sensitivity.

eess.IV↗

arXiv2024

Hardware Neural Control of CartPole and F1TENTH Race Car

Nonlinear model predictive control (NMPC) has proven to be an effective control method, but it is expensive to compute. This work demonstrates the use of hardware FPGA neural network controllers trained to imitate NMPC with supervised learning. We use these Neural Controllers (NCs) implemented on inexpensive embedded FPGA hardware for high frequency control on physical cartpole and F1TENTH race car. Our results show that the NCs match the control performance of the NMPCs in simulation and outperform it in reality, due to the faster control rate that is afforded by the quick FPGA NC inference. We demonstrate kHz control rates for a physical cartpole and offloading control to the FPGA hardware on the F1TENTH car. Code and hardware implementation for this paper are available at https:// github.com/SensorsINI/Neural-Control-Tools.

cs.RO↗

arXiv2023

Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training

Recurrent Neural Networks (RNNs) are useful in temporal sequence tasks. However, training RNNs involves dense matrix multiplications which require hardware that can support a large number of arithmetic operations and memory accesses. Implementing online training of RNNs on the edge calls for optimized algorithms for an efficient deployment on hardware. Inspired by the spiking neuron model, the Delta RNN exploits temporal sparsity during inference by skipping over the update of hidden states from those inactivated neurons whose change of activation across two timesteps is below a defined threshold. This work describes a training algorithm for Delta RNNs that exploits temporal sparsity in the backward propagation phase to reduce computational requirements for training on the edge. Due to the symmetric computation graphs of forward and backward propagation during training, the gradient computation of inactivated neurons can be skipped. Results show a reduction of $\sim$80% in matrix operations for training a 56k parameter Delta LSTM on the Fluent Speech Commands dataset with negligible accuracy loss. Logic simulations of a hardware accelerator designed for the training algorithm show 2-10X speedup in matrix computations for an activation sparsity range of 50%-90%. Additionally, we show that the proposed Delta RNN training will be useful for online incremental learning on edge devices with limited computing resources.

cs.LG↗

arXiv2023

Exploiting Alternating DVS Shot Noise Event Pair Statistics to Reduce Background Activity

Dynamic Vision Sensors (DVS) record "events" corresponding to pixel-level brightness changes, resulting in data-efficient representation of a dynamic visual scene. As DVS expand into increasingly diverse applications, non-ideal behaviors in their output under extreme sensing conditions are important to consider. Under low illumination (below ~10 lux) their output begins to be dominated by shot noise events (SNEs) which increase the data output and obscure true signal. SNE rates can be controlled to some degree by tuning circuit parameters to reduce sensitivity or temporal response bandwidth at the cost of signal loss. Alternatively, an improved understanding of SNE statistics can be leveraged to develop novel techniques for minimizing uninformative sensor output. We first explain a fundamental observation about sequential pairing of opposite polarity SNEs based on pixel circuit logic and validate our theory using DVS recordings and simulations. Finally, we derive a practical result from this new understanding and demonstrate two novel biasing techniques to reduce SNEs by 50% and 80% respectively while still retaining sensitivity and/or temporal resolution.

eess.IV↗

arXiv2023

Optimal biasing and physical limits of DVS event noise

Under dim lighting conditions, the output of Dynamic Vision Sensor (DVS) event cameras is strongly affected by noise. Photon and electron shot-noise cause a high rate of non-informative events that reduce Signal to Noise ratio. DVS noise performance depends not only on the scene illumination, but also on the user-controllable biasing of the camera. In this paper, we explore the physical limits of DVS noise, showing that the DVS photoreceptor is limited to a theoretical minimum of 2x photon shot noise, and we discuss how biasing the DVS with high photoreceptor bias and adequate source-follower bias approaches optimal noise performance. We support our conclusions with pixel-level measurements of a DAVIS346 and analysis of a theoretical pixel model.

eess.SY↗

arXiv2023

Shining light on the DVS pixel: A tutorial and discussion about biasing and optimization

The operation of the DVS event camera is controlled by the user through adjusting different bias parameters. These biases affect the response of the camera by controlling - among other parameters - the bandwidth, sensitivity, and maximum firing rate of the pixels. Besides determining the response of the camera to input signals, biases significantly impact its noise performance. Bias optimization is a multivariate process depending on the task and the scene, to which the user's knowledge about pixel design and non-idealities can be of great importance. In this paper, we go step-by-step along the signal pathway of the DVS pixel, shining light on its low-level operation and non-idealities, comparing pixel level measurements with array level measurements, and discussing and how biasing and illumination affect the pixel's behavior. With the results and discussion presented, we aim to help DVS users achieve more hardware-aware camera utilization and modelling.

eess.SY↗

arXiv2022

Measuring diameters and velocities of artificial raindrops with a neuromorphic dynamic vision sensor disdrometer

Hydrometers that can measure size and velocity distributions of precipitation are needed for research and corrections of rainfall estimates from weather radars and microwave links. Existing video disdrometers measure drop size distributions, but underestimate small raindrops and are impractical for widespread always-on IoT deployment. We propose an innovative method of measuring droplet size and velocity using a neuromorphic event camera. These dynamic vision sensors asynchronously output a sparse stream of pixel brightness changes. Droplets falling through the plane of focus create events generated by the motion of the droplet. Droplet size and speed are inferred from the stream of events. Using an improved hard disk arm actuator to reliably generate artificial raindrops, our experiments show small errors of 7% (maximum mean absolute percentage error) for droplet sizes from 0.3 to 2.5 mm and speeds from 1.3 m/s to 8.0 m/s. Each droplet requires the processing of only a few hundred to thousands of events, potentially enabling low-power always-on disdrometers that consume power proportional to the rainfall rate.

physics.ao-ph↗

arXiv2022

A 23 $μ$W Keyword Spotting IC with Ring-Oscillator-Based Time-Domain Feature Extraction

This article presents the first keyword spotting (KWS) IC which uses a ring-oscillator-based time-domain processing technique for its analog feature extractor (FEx). Its extensive usage of time-encoding schemes allows the analog audio signal to be processed in a fully time-domain manner except for the voltage-to-time conversion stage of the analog front-end. Benefiting from fundamental building blocks based on digital logic gates, it offers a better technology scalability compared to conventional voltage-domain designs. Fabricated in a 65 nm CMOS process, the prototyped KWS IC occupies 2.03mm$^{2}$ and dissipates 23 $μ$W power consumption including analog FEx and digital neural network classifier. The 16-channel time-domain FEx achieves 54.89 dB dynamic range for 16 ms frame shift size while consuming 9.3 $μ$W. The measurement result verifies that the proposed IC performs a 12-class KWS task on the Google Speech Command Dataset (GSCD) with >86% accuracy and 12.4 ms latency.

cs.AR↗

arXiv2022

Spartus: A 9.4 TOp/s FPGA-based LSTM Accelerator Exploiting Spatio-Temporal Sparsity

Long Short-Term Memory (LSTM) recurrent networks are frequently used for tasks involving time-sequential data such as speech recognition. Unlike previous LSTM accelerators that either exploit spatial weight sparsity or temporal activation sparsity, this paper proposes a new accelerator called "Spartus" that exploits spatio-temporal sparsity to achieve ultra-low latency inference. Spatial sparsity is induced using a new Column-Balanced Targeted Dropout (CBTD) structured pruning method, producing structured sparse weight matrices for a balanced workload. The pruned networks running on Spartus hardware achieve weight sparsity levels of up to 96% and 94% with negligible accuracy loss on the TIMIT and the Librispeech datasets. To induce temporal sparsity in LSTM, we extend the previous DeltaGRU method to the DeltaLSTM method. Combining spatio-temporal sparsity with CBTD and DeltaLSTM saves on weight memory access and associated arithmetic operations. The Spartus architecture is scalable and supports real-time online speech recognition when implemented on small and large FPGAs. Spartus per-sample latency for a single DeltaLSTM layer of 1024 neurons averages 1 us. Exploiting spatio-temporal sparsity on our test LSTM network using the TIMIT dataset leads to 46X speedup of Spartus over its theoretical hardware performance to achieve 9.4 TOp/s effective batch-1 throughput and 1.1 TOp/s/W power efficiency.

cs.AR↗

arXiv2022

Utility and Feasibility of a Center Surround Event Camera

Standard dynamic vision sensor (DVS) event cameras output a stream of spatially-independent log-intensity brightness change events so they cannot suppress spatial redundancy. Nearly all biological retinas use an antagonistic center-surround organization. This paper proposes a practical method of implementing a compact, energy-efficient Center Surround DVS (CSDVS) with a surround smoothing network that uses compact polysilicon resistors for lateral resistance. The paper includes behavioral simulation results for the CSDVS (see sites.google.com/view/csdvs/home). The CSDVS would significantly reduce events caused by low spatial frequencies, but amplify the informative high frequency spatiotemporal events.

cs.CV↗

arXiv2021

Bio-inspired Polarization Event Camera

The stomatopod (mantis shrimp) visual system has recently provided a blueprint for the design of paradigm-shifting polarization and multispectral imaging sensors, enabling solutions to challenging medical and remote sensing problems. However, these bioinspired sensors lack the high dynamic range (HDR) and asynchronous polarization vision capabilities of the stomatopod visual system, limiting temporal resolution to \~12 ms and dynamic range to \~ 72 dB. Here we present a novel stomatopod-inspired polarization camera which mimics the sustained and transient biological visual pathways to save power and sample data beyond the maximum Nyquist frame rate. This bio-inspired sensor simultaneously captures both synchronous intensity frames and asynchronous polarization brightness change information with sub-millisecond latencies over a million-fold range of illumination. Our PDAVIS camera is comprised of 346x260 pixels, organized in 2-by-2 macropixels, which filter the incoming light with four linear polarization filters offset by 45 degrees. Polarization information is reconstructed using both low cost and latency event-based algorithms and more accurate but slower deep neural networks. Our sensor is used to image HDR polarization scenes which vary at high speeds and to observe dynamical properties of single collagen fibers in bovine tendon under rapid cyclical loads

cs.CV↗

arXiv2021

Unraveling the paradox of intensity-dependent DVS pixel noise

Dynamic vision sensor (DVS) event camera output is affected by noise, particularly in dim lighting conditions. A theory explaining how photon and electron noise affect DVS output events has so far not been developed. Moreover, there is no clear understanding of how DVS parameters and operating conditions affect noise. There is an apparent paradox between the real noise data observed from the DVS output and the reported noise measurements of the logarithmic photoreceptor. While measurements of the logarithmic photoreceptor predict that the photoreceptor is approximately a first-order system with RMS noise voltage independent of the photocurrent, DVS output shows higher noise event rates at low light intensity. This paper unravels this paradox by showing how the DVS photoreceptor is a second-order system, and the assumption that it is first-order is generally not reasonable. As we show, at higher photocurrents, the photoreceptor amplifier dominates the frequency response, causing a drop in RMS noise voltage and noise event rate. We bring light to the noise performance of the DVS photoreceptor by presenting a theoretical explanation supported by both transistor-level simulation results and chip measurements.

eess.SY↗