{"title":"A Neuromorphic Normalization Algorithm for Stabilizing Synaptic Weights with Application to Dictionary Learning in LCA","authors":"Diego Chavez Arana, Alpha Renner, A. Sornborger","doi":"10.1145/3517343.3517357","DOIUrl":"https://doi.org/10.1145/3517343.3517357","url":null,"abstract":"Instabilities in neuromorphic machine learning can occur when synaptic updates meant to encode matrix transforms are not normalized. This phenomenon is encountered in Hebbian learning [5], where, as a synapse’s strength grows, post-synaptic activity increases, further enhancing synaptic strength, leading to a runaway condition, where synaptic strength becomes saturated [3, 7]. A number of mechanisms have been suggested for regulating and stabilizing this phenomenon [1, 2, 4, 6]. Here, we present a new neuromorphic algorithm to directly normalize a synaptic connectivity. The algorithm is based on a synfire-gated synfire chain-based information control network in concert with Hebbian synapses [8, 10]. The algorithm is designed to directly normalize synaptic weights via the minimization of a cost function. We demonstrate its effectiveness as a component of a Locally Competitive Algorithm (LCA) [9] with dictionary learning, which exhibits runaway in the absence of an effective normalization procedure [15]. LA-UR-21-31884","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127163746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Event-based dataset for classification and pose estimation","authors":"J. Turner, J. Pedersen, J. Conradt, T. Nowotny","doi":"10.1145/3517343.3517378","DOIUrl":"https://doi.org/10.1145/3517343.3517378","url":null,"abstract":"We present a new data processing pipeline for automatically labelled stereo vision data. The method is based on 3D positioning of 3D printed props with active LED markers. It provides both RGB frame and event-based data, automatically annotated with segmentation masks and ground truth position and orientation data.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133245875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Daniel Gutierrez-Galan, C. Bartolozzi, J. P. Dominguez-Morales, A. Jiménez-Fernandez, A. Linares-Barranco
{"title":"Towards the Neuromorphic Implementation of the Auditory Perception in the iCub Robotic Platform","authors":"Daniel Gutierrez-Galan, C. Bartolozzi, J. P. Dominguez-Morales, A. Jiménez-Fernandez, A. Linares-Barranco","doi":"10.1145/3517343.3517347","DOIUrl":"https://doi.org/10.1145/3517343.3517347","url":null,"abstract":"Hearing can be considered as one of the most important senses since it plays a key role in the audiovisual learning process. While a lot of effort has been made for achieving good results from the traditional approach of auditory perception, new trends such as neuromorphic computing are showing promising achievements in the implementation of brain structures for sensory perception. In this work, the design and integration of a neuromorphic event-based digital model of the auditory ascending pathway within the iCub robotic platform is proposed. This model, which comprises from the cochlea up to the inferior colliculus, replaces the traditional approach for sound processing already implemented on iCub and is able to perform sound recognition and spatial localization in real-time, allowing the implementation of auditory attention models in complex scenarios, like the cocktail party problem.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122198243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Benchmarking a Bio-inspired SNN on a Neuromorphic System","authors":"Luke Parker, Frances S. Chance, S. Cardwell","doi":"10.1145/3517343.3517365","DOIUrl":"https://doi.org/10.1145/3517343.3517365","url":null,"abstract":"INTRODUCTION Neuromorphic computing aims to derive the benefit of computational dynamics observed in biological neural systems. This has led to novel non-Von Neumann device architectures and algorithms that operate in the spiking domain. Large, densely-connected neural networks on a traditional device consume more power than their biological counterparts, especially when accounting for both training and inference. Neuromorphic devices present an opportunity to place these large, densely-connected networks on architectures that more closely resemble biological neural systems. By placing these networks in silicon, researchers in both engineering and neuroscience can better understand the cost and constraints of using neural dynamics for computation in a synthetic system. Performing computation with spikes is one of the major differences between artificial neural networks (ANNs) and spiking neural networks (SNNs). This difference is most pronounced when using a computing architecture designed to handle spiking computation such as a neuromorphic device. By implementing a SNN on both a traditional Von Neumann device, like a CPU, and a neuromorphic device, the benefits of using a native spiking architecture can be compared and analyzed. To that end, this work introduces initial findings in comparing the computational efficiency between a traditional and neuromorphic platform when implementing a bio-inspired SNN. [3] These findings contribute to the growing body of benchmark literature that highlight the performance benefits of using neuromorphic devices for bio-inspired neural network designs. The bio-inspired SNN is a simplified, spiking version of the neural network found in [3], and consists of two densely-connected layers as shown in Figure 1. The network is driven by stimulating the input layer to cause network activity, and the weights of the network are fixed as no learning occurs in the model. The neuromorphic platform chosen for this work is Intel’s Loihi, a neuromorphic chip supporting 128,000 neurons and 128 million synapses on a single chip. [4] The computation is digital, but executes using spiking dynamics inspired by neural systems. The SNN","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"547 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131788349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Felix Wang, Corinne Teeter, Sarah Luca, S. Musuvathy, Brad Aimone
{"title":"Localization through Grid-basedEncodings on Digital Elevation Models","authors":"Felix Wang, Corinne Teeter, Sarah Luca, S. Musuvathy, Brad Aimone","doi":"10.1145/3517343.3517366","DOIUrl":"https://doi.org/10.1145/3517343.3517366","url":null,"abstract":"It has been demonstrated that grid cells are encoding physical locations using hexagonally spaced, periodic phase-space representations. Theories of how the brain is decoding this phase-space representation have been developed based on neuroscience data. However, theories of how sensory information is encoded into this phase space are less certain. Here we show a method on how a navigation-relevant input space such as elevation trajectories may be mapped into a phase-space coordinate system that can be decoded using previously developed theories. Just as animals can tell where they are in a local region based on where they have been, our encoding algorithm enables the localization to a position in space by integrating measurements from a trajectory over a map. In this extended abstract, we walk through our approach with simulations using a digital elevation model.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115209875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Catherine D. Schuman, J. Plank, R. Patton, T. Potok, G. Rose
{"title":"A Framework to Enable Top-Down Co-Design of Neuromorphic Systems for Real-World Applications","authors":"Catherine D. Schuman, J. Plank, R. Patton, T. Potok, G. Rose","doi":"10.1145/3517343.3517370","DOIUrl":"https://doi.org/10.1145/3517343.3517370","url":null,"abstract":"Neuromorphic computing systems offer many attractive qualities for real-world applications, including extremely low power and relatively fast computation. There are many new and emerging devices and materials that are being evaluated by the research community for neuromorphic deployment. However, as this is still an active research field, there are relatively few application-ready, full-scale neuromorphic systems that are available to the community. As such, much of neuromorphic research is currently being performed via a bottom-up approach, in which the hardware is defined first and then applications are mapped onto the hardware after. The lack of complete neuromorphic solutions, including algorithms and software, as well as full-scale neuromorphic implementation, does not allow for application-deployment and evaluation of neuromorphic computing solutions for real-world applications. To address this issue, we have developed a complete neuromorphic framework (Figure 1), including algorithms for designing and training spiking neural networks (SNNs) for neuromorphic deployment, a common software interface to a variety of neuromorphic backends, and a field-programmable gate array (FPGA)-based neuromorphic implementation to study architectural requirements of","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117272072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Sequence Learning and Consolidation on Loihi using On-chip Plasticity","authors":"Jack W Lindsey, J. Aimone","doi":"10.1145/3517343.3517367","DOIUrl":"https://doi.org/10.1145/3517343.3517367","url":null,"abstract":"In this work we develop a model of predictive learning on neuromorphic hardware. Our model uses the on-chip plasticity capabilities of the Loihi chip to remember observed sequences of events and use this memory to generate predictions of future events in real time. Given the locality constraints of on-chip plasticity rules, generating predictions without interfering with the ongoing learning process is nontrivial. We address this challenge with a memory consolidation approach inspired by hippocampal replay. Sequence memory is stored in an initial memory module using spike-timing dependent plasticity. Later, during an offline period, memories are consolidated into a distinct prediction module. This second module is then able to represent predicted future events without interfering with the activity, and plasticity, in the first module, enabling online comparison between predictions and ground-truth observations. Our model serves as a proof-of-concept that online predictive learning models can be deployed on neuromorphic hardware with on-chip plasticity.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126857637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Efficient GPU training of LSNNs using eProp","authors":"James C. Knight, T. Nowotny","doi":"10.1145/3517343.3517346","DOIUrl":"https://doi.org/10.1145/3517343.3517346","url":null,"abstract":"Taking inspiration from machine learning libraries – where techniques such as parallel batch training minimise latency and maximise GPU occupancy – as well as our previous research on efficiently simulating Spiking Neural Networks (SNNs) on GPUs for computational neuroscience, we have extended our GeNN SNN simulator to enable spike-based machine learning research on general purpose hardware. We demonstrate that SNN classifiers implemented using GeNN and trained using the eProp learning rule can provide comparable performance to those trained using Back Propagation Through Time and show that the latency and energy usage of our SNN classifiers is up to 7 × lower than an LSTM running on the same GPU hardware.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131619407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Evaluating parameter tuning and real-time closed-loop simulation of large scale spiking networks before mapping to neuromorphic hardware: Comparing GeNN and NEST","authors":"Felix Johannes Schmitt, M. Nawrot","doi":"10.1145/3517343.3517350","DOIUrl":"https://doi.org/10.1145/3517343.3517350","url":null,"abstract":"ACM Reference Format: Felix J. Schmitt and Martin P. Nawrot. 2022. Evaluating parameter tuning and real-time closed-loop simulation of large scale spiking networks before mapping to neuromorphic hardware: Comparing GeNN and NEST. In NeuroInspired Computational Elements Conference (NICE 2022), March 28-April 1, 2022, Virtual Event, USA. ACM, New York, NY, USA, 3 pages. https://doi.org/ 10.1145/3517343.3517350","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132195147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dighanchal Banerjee, S. Dey, Arun M. George, Arijit Mukherjee
{"title":"Efficient Optimized Spike Encoding of Multivariate Time-series","authors":"Dighanchal Banerjee, S. Dey, Arun M. George, Arijit Mukherjee","doi":"10.1145/3517343.3517349","DOIUrl":"https://doi.org/10.1145/3517343.3517349","url":null,"abstract":"Spiking neural network (SNN) are emerging as a bio-plausible AI paradigm best suited for energy constrained edge use case. However the performance of SNNs largely depends upon the information content of the spike trains generated from real valued data by spike encoders - which are often found to be lossy. In this work, we have proposed a mutual information based optimisation technique of spike encoding to be used on multivariate time-series data. When tested using a spiking reservoir network, the technique is found to increase the network performance by upto 6% while performing classification task on different multivariate sensor data having variety of attributes.","PeriodicalId":208651,"journal":{"name":"Proceedings of the 2022 Annual Neuro-Inspired Computational Elements Conference","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128185595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}