arXiv subjects

Stephan Hartmann

Publications and source records attributed to Stephan Hartmann.

12 recordsLinked to original sources

arXiv2026

The SpiNNaker2 chip: a many-core platform for flexible and scalable brain-inspired computing

In deep learning, efficiency gets more and more important to compensate for the ongoing growth in model sizes and applications. Neuromorphic hardware has long been advocated as an upcoming alternative to deep networks, taking inspiration from the brain for achieving unprecedented energy efficiency. However, demonstrations of these gains only recently began to grow in complexity and real-world applicability. With SpiNNaker2, we present a chip that bridges the gap between deep networks and neuromorphic computing and allows for flexible exploration of computing approaches that combine both worlds. It features 152 processing elements equipped with an ARM M4F processor and dedicated accelerators, an extended SpiNNaker routing fabric for scalable event-based communication and a range of external interfaces for system integration, including Gbit Ethernet and an LPDDR4 memory interface. We demonstrate performance and efficiency of the SpiNNaker2 chip for neuromorphic and deep network workloads, as well as novel event-based computing approaches. For deep network workloads, the chip achieves up to 4.5 TOPS in high performance mode and up to 2.7 TOPS/W efficiency in high efficiency mode for INT8 workloads. The chip supports spiking neural networks with >150000 neurons and >1.8 billion synaptic events/s when simulated with a 1 ms time step. Its low baseline power of less than 250 mW allows for efficiency even under varying workload conditions, allowing to explore sparse and event-based modes of computation. All this demonstrates the chip's capabilities as a universal hardware platform for scalable brain-inspired computing and its combinations with mainstream deep network approaches.

cs.ET↗

arXiv2023

The Open Systems View and the Everett Interpretation

It is argued that those who defend the Everett, or `Many Worlds', interpretation of quantum mechanics should embrace what we call the general quantum theory of open systems (GT) as the proper framework in which to conduct foundational and philosophical investigation in quantum physics. GT is a wider dynamical framework than its alternative, standard quantum theory (ST). This is true even though GT makes no modifications to the quantum formalism. GT rather takes a different view, what we call the open systems view, of the formalism; i.e., in GT the dynamics of systems, whose physical states are fundamentally represented by density operators, are represented as fundamentally open as specified by an in general non-unitary dynamical map. This includes, in principle, the dynamics of the universe as a whole. We argue that the more general dynamics describable in GT can be physically motivated, that there is as much prima facie empirical support for GT as there is for ST, and that GT could be fully in the spirit of the Everett interpretation -- that there might, in short, be little reason for an Everettian not to embrace the more general theoretical landscape that GT allows one to explore.

quant-ph↗

arXiv2021

The Open Systems View

There is a deeply entrenched view in philosophy and physics, the closed systems view, according to which isolated systems are conceived of as fundamental. On this view, when a system is under the influence of its environment this is described in terms of a coupling between it and a separate system which taken together are isolated. We argue against this view, and in favor of the alternative open systems view, for which systems interacting with their environment are conceived of as fundamental, and the environment's influence is represented via the dynamical equations that govern the system's evolution. Taking quantum theories of closed and open systems as our case study, and considering three alternative notions of fundamentality: (i)~ontic fundamentality, (ii)~epistemic fundamentality, and (iii)~explanatory fundamentality, we argue that the open systems view is fundamental, and that this has important implications for the philosophy of physics, the philosophy of science, and for metaphysics.

physics.hist-ph↗

arXiv2021

The Closure of the Physical, Consciousness and Scientific Practice

We analyse the implications of the closure of the physical for experiments in the scientific study of consciousness when all the details are considered, especially how measurement results relate to physical events. It turns out that the closure of the physical has surprising implications that conflict with scientific practice. These implications point to a fundamental flaw in the paradigm underlying many experiments conducted to date and pose a challenge to any research programme that aims to ground a physical functionalist or identity-based understanding of consciousness on empirical observations.

q-bio.NC↗

arXiv2021

Hardware Implementation of an OPC UA Server for Industrial Field Devices

Industrial plants suffer from a high degree of complexity and incompatibility in their communication infrastructure, caused by a wild mix of proprietary technologies. This prevents transformation towards Industry 4.0 and the Industrial Internet of Things. Open Platform Communications Unified Architecture (OPC UA) is a standardized protocol that addresses these problems with uniform and semantic communication across all levels of the hierarchy. However, its adoption in embedded field devices, such as sensors and actors, is still lacking due to prohibitive memory and power requirements of software implementations. We have developed a dedicated hardware engine that offloads processing of the OPC UA protocol and enables realization of compact and low-power field devices with OPC UA support. As part of a proof-of-concept embedded system we have implemented this engine in a 22 nm FDSOI technology. We measured performance, power consumption, and memory footprint of our test chip and compared it with a software implementation based on open62541 and a Raspberry Pi 2B. Our OPC UA hardware engine is 50 times more energy efficient and only requires 36 KiB of memory. The complete chip consumes only 24 mW under full load, making it suitable for low-power embedded applications.

cs.AR↗

arXiv2021

The SpiNNaker 2 Processing Element Architecture for Hybrid Digital Neuromorphic Computing

This paper introduces the processing element architecture of the second generation SpiNNaker chip, implemented in 22nm FDSOI. On circuit level, the chip features adaptive body biasing for near-threshold operation, and dynamic voltage-and-frequency scaling driven by spiking activity. On system level, processing is centered around an ARM M4 core, similar to the processor-centric architecture of the first generation SpiNNaker. To speed operation of subtasks, we have added accelerators for numerical operations of both spiking (SNN) and rate based (deep) neural networks (DNN). PEs communicate via a dedicated, custom-designed network-on-chip. We present three benchmarks showing operation of the whole processor element on SNN, DNN and hybrid SNN/DNN networks.

cs.AR↗

arXiv2019

Dynamic Power Management for Neuromorphic Many-Core Systems

This work presents a dynamic power management architecture for neuromorphic many core systems such as SpiNNaker. A fast dynamic voltage and frequency scaling (DVFS) technique is presented which allows the processing elements (PE) to change their supply voltage and clock frequency individually and autonomously within less than 100 ns. This is employed by the neuromorphic simulation software flow, which defines the performance level (PL) of the PE based on the actual workload within each simulation cycle. A test chip in 28 nm SLP CMOS technology has been implemented. It includes 4 PEs which can be scaled from 0.7 V to 1.0 V with frequencies from 125 MHz to 500 MHz at three distinct PLs. By measurement of three neuromorphic benchmarks it is shown that the total PE power consumption can be reduced by 75%, with 80% baseline power reduction and a 50% reduction of energy per neuron and synapse computation, all while maintaining temporary peak system performance to achieve biological real-time operation of the system. A numerical model of this power management model is derived which allows DVFS architecture exploration for neuromorphics. The proposed technique is to be used for the second generation SpiNNaker neuromorphic many core system.

cs.NE↗

arXiv2017

Pattern representation and recognition with accelerated analog neuromorphic systems

Despite being originally inspired by the central nervous system, artificial neural networks have diverged from their biological archetypes as they have been remodeled to fit particular tasks. In this paper, we review several possibilites to reverse map these architectures to biologically more realistic spiking networks with the aim of emulating them on fast, low-power neuromorphic hardware. Since many of these devices employ analog components, which cannot be perfectly controlled, finding ways to compensate for the resulting effects represents a key challenge. Here, we discuss three different strategies to address this problem: the addition of auxiliary network components for stabilizing activity, the utilization of inherently robust architectures and a training method for hardware-emulated networks that functions without perfect knowledge of the system's dynamics and parameters. For all three scenarios, we corroborate our theoretical considerations with experimental results on accelerated analog neuromorphic platforms.

q-bio.NC↗

arXiv2017

Neuromorphic Hardware In The Loop: Training a Deep Spiking Network on the BrainScaleS Wafer-Scale System

Emulating spiking neural networks on analog neuromorphic hardware offers several advantages over simulating them on conventional computers, particularly in terms of speed and energy consumption. However, this usually comes at the cost of reduced control over the dynamics of the emulated networks. In this paper, we demonstrate how iterative training of a hardware-emulated network can compensate for anomalies induced by the analog substrate. We first convert a deep neural network trained in software to a spiking network on the BrainScaleS wafer-scale neuromorphic system, thereby enabling an acceleration factor of 10 000 compared to the biological time domain. This mapping is followed by the in-the-loop training, where in each training step, the network activity is first recorded in hardware and then used to compute the parameter updates in software via backpropagation. An essential finding is that the parameter updates do not have to be precise, but only need to approximately follow the correct gradient, which simplifies the computation of updates. Using this approach, after only several tens of iterations, the spiking network shows an accuracy close to the ideal software-emulated prototype. The presented techniques show that deep spiking networks emulated on analog neuromorphic devices can attain good computational performance despite the inherent variations of the analog substrate.

cs.NE↗

arXiv2015

Hawking Radiation and Analogue Experiments: A Bayesian Analysis

We present a Bayesian analysis of the epistemology of analogue experiments with particular reference to Hawking radiation. First, we prove that such experiments can be confirmatory in Bayesian terms based upon appeal to 'universality arguments'. Second, we provide a formal model for the scaling behaviour of the confirmation measure for multiple distinct realisations of the analogue system and isolate a generic saturation feature. Finally, we demonstrate that different potential analogue realisations could provide different levels of confirmation. Our results provide a basis both to formalise the epistemic value of analogue experiments that have been conducted and to advise scientists as to the respective epistemic value of future analogue experiments.

physics.hist-ph↗

arXiv2012

Generalized Dicke States

Quantum master equations are an important tool in quantum optics and quantum information theory. For systems comprising a small to medium number of atoms (or qubits), the non-truncated equations are usually solved numerically. In this paper, we present a group-theoretical superoperator method that helps solving these equations. To do so, we exploit the SU(4)-symmetry of the respective Lindblad operator and construct basis states that generalize the well-known Dicke states. This allows us to solve various problems analytically and to considerably reduce the complexity of problems that can only be solved numerically. Finally, we present three examples that illustrate the proposed method.

quant-ph↗

arXiv2000

Coherence, Belief Expansion and Bayesian Networks

We construct a probabilistic coherence measure for information sets which determines a partial coherence ordering. This measure is applied in constructing a criterion for expanding our beliefs in the face of new information. A number of idealizations are being made which can be relaxed by an appeal to Bayesian Networks.

cs.AI↗