25 papers · ranked by Valyu relevance
Serpen, Gursel
We are proposing fully parallel and maximally distributed hardware realization of a generic neurocomputing system. More specifically, the proposal relates to the wireless sensor networks technology to serve as a massively parallel and fully distributed hardware platform to implement and realize artificial neural…
Mingjing Li, Huihui Zhou, Xiaofeng Xu, Zhiwei Zhong + 15 more
There is a growing necessity for edge training to adapt to dynamically changing environments. Neuromorphic computing represents a significant pathway for highly efficient intelligent computation in energy-constrained edges, but existing neuromorphic architectures lack the ability of directly training spiking neural…
Afnan M. Alhassan, Nouf I. Altmami, Choon Chiat Oh
Background/Objectives: Globally, one of the most dreadful and rapidly spreading illnesses is skin cancer, and it is acknowledged as a lethal form of cancer due to the abnormal growth of skin cells. Mostly, classifying and diagnosing the types of skin lesions is complex, and recognizing tumors from dermoscopic images…
Matteo De Matola, Giorgio Arcara
Convolutional neural networks (CNNs) are a class of artificial neural networks (ANNs). Since the early 2010s, they have been widely adopted as models of primate vision and classifiers of neuroimaging data, becoming relevant for a wealth of neuroscientific fields. However, the majority of neuroscience researchers come…
Benedetto Leto, A.R. De Pierro, Noah Pacik-Nelson, Korneel Van den Berghe + 3 more
—Neuromorphic accelerators offer promising platforms for machine learning (ML) inference by leveraging eventdriven, spatially-expanded architectures that naturally exploit unstructured sparsity through co-located memory and compute. However, their unique architectural characteristics create performance dynamics that…
Sander Liessem, Samuel K. Asinof, Aljoscha Nern, Marissa Sumathipala + 5 more
Nervous systems can process information in serial or in parallel, trading off efficiency for flexibility and speed. How these network architectures are implemented across sensorimotor pathways to control behavior is unclear. We investigate this tradeoff directly in Drosophila by comparing neuronal circuits underlying…
Jiahao Li, Ming Xu, Heng Dong, Bin Lan + 5 more
The deployment of Spiking Neural Networks (SNNs) on resource-constrained edge devices is hindered by a critical algorithm-hardware mismatch: a fundamental trade-off between the accuracy degradation caused by aggressive quantization and the resource redundancy stemming from traditional decoupled hardware designs. To…
Kangbo Bai, Zhantong Zhu, Yifan Ding, Tianyu Jia
In large-scale distributed LLM training, communication between devices becomes the key performance bottleneck. Chiplet technology can integrate multiple dies into a package to scale-up node performance with higher bandwidth. Meanwhile, optical interconnect (OI) technology offers long-reach, high-bandwidth links, making…
Authors not listed
Machine Learning Interatomic Potentials (MLIPs), trained with Quantum Mechanics data, can model potential energy surfaces for molecular systems with very high accuracy and extreme speedups compared to reference quantum calculations, offering a powerful tool for studying complex chemical and biological systems. This…
Pierpaolo Perticaroli, Roberto Ammendola, Andrea Biagioni, Ottorino Frezza + 9 more
Spiking neural networks (SNNs) run today on a fragmented landscape of hardware: dedicated neuromorphic processors, application-specific FPGA accelerators, and large-scale neuroscience simulators, each typically built around a fixed neuron model, execution strategy, or workload class. We present AIGOR, a modular…
Andrew Polar, Michael Poluektov
Training on disjoint datasets can serve two primary goals: accelerating data processing and enabling federated learning. It has already been established that Kolmogorov-Arnold networks (KANs) are particularly well suited for federated learning and can be merged through simple parameter averaging. While the federated…
Kwangjun Lee, Lorenzo Baracco, Cyriel M.A. Pennartz, Mototaka Suzuki + 1 more
Artificial neural networks commonly have deep hierarchical structures that were originally inspired by the neuroanatomical evidence of cortico-cortical connectivity pattern found in the mammalian brain. Largely under-represented in those models are non-hierarchical aspects of brain architecture, namely the subcortical…
Yuma Osako, Aineias Arango, Toshitake Asabuki
Animals flexibly combine learned behaviors into novel actions without practicing their combinations, yet the computational mechanisms that enable independently acquired computations to be expressed in parallel remain unclear. Here we show that feedback geometry during learning determines whether recurrent dynamics can…
Shanmuga Venkatachalam, Prabhu Vellaisamy, Harideep Nair, Wei-Che Huang + 4 more
—Leading experts from both communities have suggested the need to (re)connect research in neuroscience and artificial intelligence (AI) to accelerate the development of nextgeneration AI innovations. They term this convergence as NeuroAI. Previous research has established temporal neural networks (TNNs) as a promising…
Francisco Requena-Domínguez, Rafaela Benítez-Rochel, Ezequiel López-Rubio
The dynamics of the original Hopfield network is asynchronous (sequential) (updates the state of only one neuron per time step). In this paper, we propose a new tool and a new dynamics to reduce the processing time by updating one or more neurons simultaneously per instant while ensuring process convergence and aiming…
Authors not listed
Molecular dynamics (MD) is a powerful tool for exploring the behavior of atomistic systems, but its reliance on sequential numerical integration limits simulation efficiency. We present MDtrajNet-1, a foundational AI model that directly generates MD trajectories across chemical space, bypassing force calculations and…
Xundong Wu, Pengfei Zhao, Zilin Yu, Lei Ma + 6 more
Title: Summary Why have modern artificial neural networks not adopted the nonlinear dendritic structures found in biological brain cells, and what is the core advantage of such active dendritic units? While early studies suggested that dendritic nonlinearities can enhance learning capabilities by boosting capacity, we…
Karla Ivankovic, Anastasios Dimou, Justo Montoya-Gálvez, Riccardo Zucca + 2 more
Understanding how the brain represents information is a central challenge in neuroscience and a practical bottleneck for brain-computer interfaces. Existing analytical tools cannot identify neural representations directly from neural activity data. We introduce MultiPEC, a data-driven method that discovers neural…
Authors not listed
Recent advances in machine learning force fields (MLFF) have significantly extended the reach of atomistic simulations. Continuous progress in this field requires reliable reference datasets, accurate MLFF architectures, and efficient active learning strategies to enable robust modeling of complex molecular and…
Jie Lin, Mai Lu
The anti-interference characteristics of the neural network have a key impact on its information processing ability in complex environments. Most of the existing research focuses on small-scale networks and simplified models, and there is still a lack of systematic discussion on the influence mechanism of large-scale…
G. Kandemir, D. H. Duncan, D. van Moorselaar, J. Theeuwes
For almost half a century, target-distractor similarity has been known to induce different visual search modes. When a target is highly salient, it can pop out, suggesting parallel processing of all items irrespective of set size. By contrast, high similarity among items requires item-by-item comparison with an…
Authors not listed
Machine olfaction—the artificial replication of the sense of smell—faces significant challenges due to the absence of large, standardized training datasets. Unlike vision, language, and audio models, which benefit from extensive corpora such as ImageNet, GLUE, and AudioSet, olfaction lacks scaled equivalents and…
Chenggang Chen, Zhiyu Yang, Xiaoqin Wang
Whether the auditory cortex has parallel pathways for sound identification (“what”) and localization (“where”), and whether it contains a map of auditory space, is debated. Here, we examined the low-dimensional structure (manifold) of “what” and “where” representations in deep neural network models of sound…
Pavel Tolmachev, Tatiana A. Engel
Trained recurrent neural networks (RNNs) have become the leading framework for modelling neural dynamics in the brain, owing to their capacity to mimic how population-level computations arise from interactions among many units with heterogeneous responses. RNN units are commonly modelled using various nonlinear…
Authors not listed
Bayesian optimization (BO) has become increasingly important for experimental optimization across scientific domains, yet implementing BO pipelines requires significant programming expertise and familiarity with specialized frameworks. This creates a barrier for domain experts who could benefit from BO but lack the…