Neural Computation最新文献_第2页

Toward Enhancing RMSProp With Forward-Looking Gradient Updates for Complex Loss Landscapes. 增强RMSProp与前瞻性梯度更新复杂的损失景观。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-04-07 DOI: 10.1162/NECO.a.1514

Rafał Wolniak, Bożena Kostek

{"title":"Toward Enhancing RMSProp With Forward-Looking Gradient Updates for Complex Loss Landscapes.","authors":"Rafał Wolniak, Bożena Kostek","doi":"10.1162/NECO.a.1514","DOIUrl":"https://doi.org/10.1162/NECO.a.1514","url":null,"abstract":"This letter introduces a novel algorithm for training deep neural networks with many nonlinear layers. Our method uses an approximated integrated gradient that is averaged over the range of the weight update to more accurately capture the loss change resulting from parameter updates. Unlike standard gradients, this average gradient improves learning efficiency in certain scenarios. We incorporate the approximated average gradient into RMSProp and compare the resulting algorithm to conventional RMSProp and Adam. We evaluate the approach on deep models lacking skip connections, such as those with many nonlinear activations and no residual structure, where traditional methods typically encounter difficulties. These models that focus on extracting high-order features create a loss landscape more akin to that of a biological brain. Our method requires significantly fewer iterations to reach a target training loss on MNIST, Fashion MNIST, and IMDb benchmarks for both convolutional and fully connected architectures across different initialization schemes. While our approach incurs moderately higher computational and memory costs compared to standard RMSProp, its performance on shallow models remains comparable. Nevertheless, our main contributions are (1) introducing the average gradient concept as an efficient alternative to computing high-order derivatives, (2) offering a novel factorization formula for approximating the average gradient, accompanied by a formal derivation., and (3) showing an example algorithm that leverages this formula to enhance the efficiency of RMSProp for some models, as validated by our evaluation.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":" ","pages":"1-29"},"PeriodicalIF":2.1,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147635089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Implicit Generative Modeling by Kernel Similarity Matching. 基于核相似度匹配的隐式生成建模。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-04-07 DOI: 10.1162/NECO.a.1511

Shubham Choudhary, Paul Masset, Demba Ba

{"title":"Implicit Generative Modeling by Kernel Similarity Matching.","authors":"Shubham Choudhary, Paul Masset, Demba Ba","doi":"10.1162/NECO.a.1511","DOIUrl":"https://doi.org/10.1162/NECO.a.1511","url":null,"abstract":"Understanding how the brain encodes stimuli has been a fundamental problem in computational neuroscience. Insights into this problem have led to the design and development of artificial neural networks that learn representations by incorporating brain-like learning abilities. Recently, learning representations by capturing similarity among input samples has been studied (Pehlevan et al., 2018) to tackle this problem. This approach, however, has thus far been used only to learn downstream features from an input and has not been studied in the context of a generative paradigm, where one can map the representations back to the input space, incorporating not only bottom-up interactions (stimuli→latent) but also learning features in a top-down manner (latent→stimuli). We investigate a kernel similarity matching framework for generative modeling. Starting with a modified sparse coding objective for learning representations proposed in prior work (Olshausen & Field, 1996; Tolooshams & Ba, 2021), we demonstrate that representation learning in this context is equivalent to maximizing similarity between the input kernel and a latent kernel. We show that an implicit generative model arises from learning the kernel structure in the latent space and show how the framework can be adapted to learn manifold structures, potentially providing insights as to how task representations can be encoded in the brain. To solve the objective, we propose a novel alternate direction method of multipliers (ADMM)-based algorithm and discuss the interpretation of the optimization process. Finally, we discuss how this representation learning problem can lead toward a biologically plausible architecture to learn the model parameters that ties together representation learning using similarity matching (a bottom-up approach) with predictive coding (a top-down approach).","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":" ","pages":"1-55"},"PeriodicalIF":2.1,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147634986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Thousand Brains Systems: Sensorimotor Intelligence for Rapid, Robust Learning and Inference. 千脑系统：快速、稳健学习和推理的感觉运动智能。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-04-07 DOI: 10.1162/NECO.a.1508

Niels Leadholm, Viviane Clay, Scott Knudstrup, Hojae Lee, Jeff Hawkins

{"title":"Thousand Brains Systems: Sensorimotor Intelligence for Rapid, Robust Learning and Inference.","authors":"Niels Leadholm, Viviane Clay, Scott Knudstrup, Hojae Lee, Jeff Hawkins","doi":"10.1162/NECO.a.1508","DOIUrl":"https://doi.org/10.1162/NECO.a.1508","url":null,"abstract":"Current AI systems achieve impressive performance on many tasks, yet they lack core attributes of biological intelligence, including rapid, continual learning, representations grounded in sensorimotor interactions, and structured knowledge that enables efficient generalization. Neuroscience theory suggests that mammals evolved flexible intelligence through the replication of a semi-independent, sensorimotor module, a functional unit known as a cortical column. To address the disparity between biological and artificial intelligence, Thousand Brains systems were proposed as a means of mirroring the architecture of cortical columns and their interactions. In our current work, we evaluate the unique properties of Monty, the first implementation of a Thousand Brains system. We focus on 3D object perception and, in particular, the combined task of object recognition and pose estimation. Utilizing the YCB data set of household objects, we first assess Monty's use of sensorimotor learning to build structured representations, finding that these enable robust generalization. These representations include an emphasis on classifying objects by their global shape, as well as a natural ability to detect object symmetries. We then explore Monty's use of model-free and model-based policies to enable rapid inference by supporting principled movements. We find that such policies complement Monty's modular architecture, a design that can accommodate communication between modules to further accelerate inference speed via a novel voting algorithm. Finally, we examine Monty's use of associative, Hebbian-like binding to enable rapid, continual, and computationally efficient learning, properties that compare favorably to current deep learning architectures. While Monty is still in a nascent stage of development, these findings support Thousand Brains systems as a powerful and promising new approach to AI and reinforce the importance of sensorimotor learning for developing intelligent systems.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":" ","pages":"1-52"},"PeriodicalIF":2.1,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147635088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multiclass Linear Perceptrons With Multiplicative Margins 具有乘边的多类线性感知器。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1502

Dmitri Rachkovskij;Evgeny Osipov;Olexander Volkov;Daswin De Silva;Denis Kleyko

{"title":"Multiclass Linear Perceptrons With Multiplicative Margins","authors":"Dmitri Rachkovskij;Evgeny Osipov;Olexander Volkov;Daswin De Silva;Denis Kleyko","doi":"10.1162/NECO.a.1502","DOIUrl":"10.1162/NECO.a.1502","url":null,"abstract":"This article introduces a family of multiclass linear perceptron classifiers with a multiplicative margin mechanism (MMPerc), as an alternative to standard margin-free and additive margin perceptrons. The multiplicative formulation enforces classification confidence by requiring the true class score to exceed that of competing classes by a specified fraction of itself rather than by a fixed additive threshold. This avoids dependence on score magnitudes arising from varied norms of data and class weight vectors. We propose several architectural and algorithmic variants of MMPerc, derive associated loss functions and mistake bounds for both linearly separable and nonseparable data, and analyze key design considerations, including bias, margin threshold selection, and training modes. Extensive experiments on synthetic and real data sets show that MMPerc classifiers typically outperform the standard perceptron, as well as classic baselines such as support vector machines and ridge classifiers. Owing to their simplicity, minimalistic design, and computational efficiency, MMPerc classifiers are promising candidates for conventional machine learning tasks, linear evaluation of deep neural networks, integration with hyperdimensional computing and vector symbolic architecture representations, and deployment in resource-constrained applications.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":"38 4","pages":"602-650"},"PeriodicalIF":2.1,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147366897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Inhibitory Feedback Enables Predictive Learning of Multiple Sequences in Neural Networks 抑制反馈使神经网络中多序列的预测学习成为可能。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1504

Matteo Saponati;Martin Vinck

引用次数: 0

The Cooperative Network Architecture: Learning Structured Networks as Representation of Sensory Patterns 合作网络架构：作为感觉模式表征的学习结构化网络。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1505

Pascal J. Sager;Jan M. Deriu;Benjamin F. Grewe;Thilo Stadelmann;Christoph von der Malsburg

引用次数: 0

Force Learning in Balanced Cortical E-I Networks 平衡皮质E-I网络的力学习。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1503

Takashi Kanamaru;Kazuyuki Aihara

{"title":"Force Learning in Balanced Cortical E-I Networks","authors":"Takashi Kanamaru;Kazuyuki Aihara","doi":"10.1162/NECO.a.1503","DOIUrl":"10.1162/NECO.a.1503","url":null,"abstract":"Force learning is a learning method for generating various types of complex dynamics in recurrent neural networks (RNNs), which is related to the reservoir computing (RC). RC uses an RNN called reservoir whose synaptic weights are randomly generated and fixed during learning. Force learning trains these synaptic weights inside the reservoir networks. Although force learning can be used as an effective tool for machine learning, possibilities of its realization in the brain are not often discussed. Here, in order to consider the possibilities of its realization in the brain, force learning is applied to an excitatory and inhibitory (E-I) network that models the cerebral cortex. A multimodule network composed of excitatory and inhibitory neurons is defined, and a readout is put outside, similar to a conventional reservoir. The output of this network is calculated at the readout as a linear combination of the filtered average firing rates of the excitatory neurons in the modules. Feedback connections that provide output back to the excitatory neurons in the modules with random strength are also added to this network. This network typically shows transitive chaotic synchronization, in which synchronizing modules are rearranged chaotically and intermittently. Under such conditions, our E-I network is trained to generate sinusoidal periodic signals for simplicity with force learning. When adjusting the E-I activity, it is observed that the efficiency of force learning is maximized at an optimal E-I balance near an edge of chaos. These results imply that the cooperation of excitatory and inhibitory neurons is required when force learning works effectively in the brain, although usual reservoir networks don’t distinguish these two kinds of neurons.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":"38 4","pages":"573-601"},"PeriodicalIF":2.1,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11455132","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147366928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Potential for Reinforcement Learning in the Cerebellum 小脑强化学习的潜力。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1507

Richard W. Prager;Richard Apps

{"title":"Potential for Reinforcement Learning in the Cerebellum","authors":"Richard W. Prager;Richard Apps","doi":"10.1162/NECO.a.1507","DOIUrl":"10.1162/NECO.a.1507","url":null,"abstract":"This article explores how simple reinforcement learning algorithms might be implemented by the anatomy of the cerebellum. In doing this, we highlight which anatomical and physiological details are most important for assessing algorithmic fit, and we discuss which algorithm components are easiest to accommodate in a neural system. We describe hypothetical cerebellar implementations of four reinforcement learning algorithms and discuss the anatomical plausibility of the various components required. We show how one of the algorithms can learn to generate short sequences of actions without continuous information on the resulting changes to the environment. We finish with simulations that illustrate the way that the algorithms learn to solve the problem of balancing an inverted pendulum, commonly known as the cart-pole problem. We highlight two physiological features: reward signals and combining information across time, that indicate that some sort of reinforcement learning adaptation may be taking place. We also describe why the commonly used algorithmic feature, an eligibility trace, presents particular problems to implement in known neural anatomy.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":"38 4","pages":"499-537"},"PeriodicalIF":2.1,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11455135","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147367212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ReBaCCA-ss: Relevance-Balanced Continuum Correlation Analysis With Smoothing and Surrogating for Quantifying Similarity Between Population Spiking Activities ReBaCCA-ss：基于平滑和代入的关联平衡连续体相关分析

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-17 Epub Date: 2026-03-24 DOI: 10.1162/NECO.a.1501

Xiang Zhang;Chenlin Xu;Zhouxiao Lu;Haonan Wang;Dong Song

{"title":"ReBaCCA-ss: Relevance-Balanced Continuum Correlation Analysis With Smoothing and Surrogating for Quantifying Similarity Between Population Spiking Activities","authors":"Xiang Zhang;Chenlin Xu;Zhouxiao Lu;Haonan Wang;Dong Song","doi":"10.1162/NECO.a.1501","DOIUrl":"10.1162/NECO.a.1501","url":null,"abstract":"Quantifying similarity between population spike patterns is essential for understanding how neural dynamics encode information. Traditional approaches, which combine kernel smoothing, principal component analysis, and canonical correlation analysis (CCA), have limitations: smoothing kernel bandwidths are often empirically chosen, CCA maximizes alignment between patterns without considering the variance explained within patterns, and baseline correlations from stochastic spiking are rarely corrected. We introduce ReBaCCA-ss (relevance-balanced continuum correlation analysis with smoothing and surrogating), a novel framework that addresses these challenges through three innovations: (1) balancing alignment and variance explanation via continuum canonical correlation, (2) correcting for noise using surrogate spike trains, and (3) selecting the optimal kernel bandwidth by maximizing the difference between true and surrogate correlations. ReBaCCA-ss is validated on both simulated data and hippocampal recordings from rats performing a delayed nonmatch-to-sample task. It reliably identifies spatiotemporal similarities between spike patterns. Combined with multidimensional scaling, ReBaCCA-ss reveals structured neural representations across trials, events, sessions, and animals, offering a powerful tool for neural population analysis.","PeriodicalId":54731,"journal":{"name":"Neural Computation","volume":"38 4","pages":"651-680"},"PeriodicalIF":2.1,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147367240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Perceptual Processes as Charting Operators. 作为图表运算符的感知过程。

IF 2.1 4区计算机科学

Neural Computation Pub Date : 2026-03-05 DOI: 10.1162/NECO.a.1506

Peter Neri

引用次数: 0