Review
Cell Biology
Ben Jiwon Choi, Yu-Chieh David Chen, Claude Desplan
Summary: During the development of the vertebrate nervous systems, genetic programs assemble an immature circuit that is subsequently refined by neuronal activity evoked by external stimuli. Prior to sensory experience, the developing nervous system also triggers correlated network-level neuronal activity, with retinal waves in the developing vertebrate retina being one example. Drosophila is emerging as a model for studying correlated spontaneous activity in the visual system, comparing it with mammals.
GENES & DEVELOPMENT
(2021)
Article
Multidisciplinary Sciences
Jacqueline Bloch, Andrea Cavalleri, Victor Galitski, Mohammad Hafezi, Angel Rubio
Summary: A goal of modern condensed-matter physics is to search for states of matter with emergent properties and desirable functionalities. By controlling light-matter interactions, it is possible to manipulate and synthesize strongly correlated quantum matter, leading to phenomena like photon-mediated superconductivity, cavity fractional quantum Hall physics, and optically driven topological phenomena.
Review
Neurosciences
Rava Azeredo da Silveira, Fred Rieke
Summary: Neurons in the brain represent information through their collective activity, and the fidelity of neural population code depends on how variability in neuron responses is shared. Two decades of studies have shown the importance of noise correlations in neural coding, with theoretical developments and new approaches providing valuable insights. A geometrical picture of how noise correlations impact the neural code is emphasized in this review.
ANNUAL REVIEW OF NEUROSCIENCE, VOL 44, 2021
(2021)
Article
Optics
Zhan-Ming Li, Shi-Bao Wu, Jun Gao, Heng Zhou, Zeng-Quan Yan, Ruo-Jing Ren, Si-Yuan Yin, Xian-Min Jin
Summary: Quantum imaging using photon pairs with strong correlations has been applied in various fields, enabling the building of photon-limited images in low-light conditions. Deep learning optimization algorithms efficiently solve inverse imaging problems related to shot noise and background noise. This research pushes low-light imaging techniques to the single-photon level in real time, allowing for deep-learning-enhanced quantum imaging.
Article
Physics, Multidisciplinary
Jin-Jian Zhou, Jinsoo Park, Iurii Timrov, Andrea Floris, Matteo Cococcioni, Nicola Marzari, Marco Bernardi
Summary: Electron-phonon interactions play a crucial role in condensed matter, governing various phenomena in materials. Density functional theory often fails to accurately describe these interactions in correlated electron systems. By utilizing Hubbard-corrected density functional theory and its linear response extension, accurate calculations of electron-phonon interactions in a wide range of materials can be achieved.
PHYSICAL REVIEW LETTERS
(2021)
Article
Quantum Science & Technology
Simon J. D. Phoenix, Faisal Shah Khan, Berihu Teklu
Summary: The production and manipulation of quantum correlation protocols is an active area of research, where the quantum nature of the correlation can be used to achieve properties unattainable in a classical framework. This work focuses on measuring the strength of correlation between quantum systems, with a special emphasis on multipartite systems.
QUANTUM INFORMATION PROCESSING
(2021)
Article
Computer Science, Information Systems
Geon Kang, Seung-Chan Kim
Summary: In this study, a novel method for learning and recognizing material properties using self-emitted acoustic signals and mimicking acoustic recognition mechanisms found in animals is presented. The proposed approach, based on convolutional neural networks, achieved high accuracies in recognizing texture and density information compared to conventional machine learning methods. Experimental validation was conducted to demonstrate the effectiveness of the method and its potential for incorporating acoustic recognition functionality into computers.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING
(2022)
Article
Biology
Sacha Sokoloski, Amir Aschner, Ruben Coen-Cagli
Summary: This paper proposes a response model based on mixture models and exponential families, which can capture the variability and covariability in large-scale neural recordings. Additionally, the model facilitates accurate Bayesian decoding, provides a closed-form expression for the Fisher information, and is compatible with theories of probabilistic population coding.
Article
Materials Science, Multidisciplinary
Xiao-Yong Feng, Haijie Cai, Jianhui Dai
Summary: The study reveals that under certain conditions, the ideal half-semimetal feature can be realized in correlated electron systems, exhibiting unique magnetic properties and phase transition behavior.
Article
Materials Science, Multidisciplinary
Olivier Simard, Shintaro Takayoshi, Philipp Werner
Summary: This paper discusses the optical conductivity of correlated electron systems, focusing on the nature and effect of the most relevant vertex corrections in finite-dimensional systems.
Article
Telecommunications
Kyuhyuk Chung
Summary: This paper investigates the non-orthogonal multiple access (NOMA) scheme for correlated information sources (CIS), proposing a non-SIC NOMA scheme that allows all users to receive the common information. The data rate achieved by this non-SIC NOMA scheme is derived and the achievable rate region is analyzed. Numerical results confirm the effectiveness of the proposed non-SIC NOMA scheme compared to the conventional SIC NOMA scheme.
IEEE COMMUNICATIONS LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Konstantin Posch, Juergen Pilz
Summary: This article presents a novel approach for training deep neural networks using Bayesian techniques, which allows for easy evaluation of model uncertainty and robustness to overfitting. The proposed method outperforms other Bayesian methods in terms of predictive accuracy and uncertainty estimation.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Review
Chemistry, Multidisciplinary
Roman Kuzian
Summary: The discovery of high-Tc superconductivity in cuprates in 1986 shifted the focus of solid-state research to strongly correlated transition metal compounds, which had previously been considered exotic worlds only of interest to pure theorists. Condensed matter physics topics such as high-Tc superconductivity, colossal magnetoresistance, multiferroicity, and ferromagnetism in diluted magnetic semiconductors are often related to these strongly correlated systems. The study of these compounds requires methods and models beyond the mean field approximation, and examples of response function calculations are discussed for the interpretation of experimental information.
Article
Chemistry, Physical
Haodong Zhang, Jingxiang Zou, Xiaochuan Ren, Shuhua Li
Summary: The Anequation-of-motion block-correlated coupled cluster method based on the generalized valence bond wave function (EOM-GVB-BCCC) is proposed to describe low-lying excited states for strongly correlated systems. The EOM-GVB-BCCC2b method, which includes up to two-pair correlation, has been successfully implemented and tested for several strongly correlated systems. The results of EOM-GVB-BCCC2b for a water hexamer and four conjugated diradical species are consistent with the density matrix renormalization group (DMRG) results.
JOURNAL OF PHYSICAL CHEMISTRY LETTERS
(2023)
Article
Materials Science, Multidisciplinary
Thomas Hansen, Lars Bojer Madsen
Summary: This study investigates high-harmonic generation from doped, correlated materials using the one-dimensional Hubbard model. It finds that doping has little to no effect on the dynamics and spectra for small electron-electron correlation, but has a significant impact for larger correlation. The findings are explained through a quasiparticle-based picture.
Article
Physics, Multidisciplinary
Christian Keup, Tobias Kuehn, David Dahmen, Moritz Helias
Summary: This study explores the differences in chaos phenomena between rate and binary networks, finding qualitative distinctions in dynamics, chaotic characteristics, dimensionality expansion, etc.
Article
Physics, Multidisciplinary
Alexander van Meegen, Tobias Kuehn, Moritz Helias
Summary: This study unifies the field-theoretical approach to neuronal networks with large deviations theory, deriving a rate function resembling Kullback-Leibler divergence through field theory to enable data-driven inference of model parameters and calculation of fluctuations beyond mean-field theory. Additionally, the study reveals a regime with fluctuation-induced transitions between mean-field solutions.
PHYSICAL REVIEW LETTERS
(2021)
Article
Neurosciences
Stefan Dasbach, Tom Tetzlaff, Markus Diesmann, Johanna Senk
Summary: This study explores the effects of limited synaptic weight resolution on the dynamics of spiking neuronal networks, finding that a naive discretization may distort spike-train statistics, but preserving the mean and variance of total synaptic input currents can maintain firing statistics for certain network types. Even with a discretization of synaptic weights, substantial deviations in firing statistics may occur, emphasizing the importance of careful validation and preservation of specific network characteristics.
FRONTIERS IN NEUROSCIENCE
(2021)
Article
Neurosciences
Arne Heittmann, Georgia Psychou, Guido Trensch, Charles E. Cox, Winfried W. Wilcke, Markus Diesmann, Tobias G. Noll
Summary: This article utilizes the new IBM INC-3000 prototype FPGA-based neural supercomputer to implement a popular model of the cortical microcircuit, achieving a high speed-up factor and demonstrating the potential of FPGA systems for neural modeling.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Neurosciences
Sebastian Spreizer, Johanna Senk, Stefan Rotter, Markus Diesmann, Benjamin Weyers
Summary: The article introduces a web-based Graphical User Interface (GUI) for simulation software of spiking neuronal network models, reducing the entry barrier for students and early career scientists in computational neuroscience. The NEST Desktop tool, with graphical elements for creating, configuring, running, and analyzing network models, enhances the quality and intensity of teaching in computational neuroscience. The availability of this tool on public resources contributes to equal opportunities in the field.
Article
Physics, Multidisciplinary
Lorenzo Tiberi, Jonas Stapmanns, Tobias Kuehn, Thomas Luu, David Dahmen, Moritz Helias
Summary: Criticality is closely linked to optimal computational capacity, but the lack of a renormalized theory of critical brain dynamics hinders insights into biological information processing. A renormalized theory of a prototypical neural field theory is presented, focusing on the flow of couplings across different length scales to achieve an effective trade-off between linearity and nonlinearity for information storage and computation.
PHYSICAL REVIEW LETTERS
(2022)
Article
Biochemical Research Methods
Younes Bouhadjar, Dirk J. Wouters, Markus Diesmann, Tom J. Tetzlaff
Summary: This study proposes a biologically-based algorithm for sequence learning, prediction, and replay. The algorithm can continuously learn complex sequences in an unsupervised manner. The study also sheds light on the mechanisms of sequence processing speed and provides an explanation for the observed fast sequence replay in the hippocampus and neocortex.
PLOS COMPUTATIONAL BIOLOGY
(2022)
Article
Computer Science, Theory & Methods
J. Pronold, J. Jordan, B. J. N. Wylie, I Kitayama, M. Diesmann, S. Kunkel
Summary: Simulation is an important approach for studying complex dynamic systems, especially in the field of biological neural networks. This study demonstrates the effectiveness of common techniques in improving simulation accuracy and efficiency.
PARALLEL COMPUTING
(2022)
Article
Biochemical Research Methods
Johanna Senk, Birgit Kriener, Mikael Djurfeldt, Nicole Voges, Han-Jia Jiang, Lisa Schuettler, Gabriele Gramelsberger, Markus Diesmann, Hans E. Plesser, Sacha J. van Albada
Summary: Sustainable research on computational models of neuronal networks requires understandable, reproducible, and extendable published models. However, missing details or ambiguities in mathematical concepts, algorithmic implementations, or parameterizations hinder progress. This work aims to provide complete and concise descriptions of network connectivity and guide the implementation of connection routines in simulation software and neuromorphic hardware systems. A review of existing models reveals a substantial proportion of ambiguous descriptions. Based on this review, a set of connectivity concepts is derived and a unified graphical notation is proposed to facilitate intuitive understanding of network properties. The proposed standardizations are expected to contribute to unambiguous descriptions and reproducible implementations of neuronal network connectivity in computational neuroscience.
PLOS COMPUTATIONAL BIOLOGY
(2022)
Article
Biochemical Research Methods
Younes S. Bouhadjar, Dirk J. Wouters, Markus S. Diesmann, Tom Tetzlaff
Summary: Humans and other animals benefit from exploring multiple alternative solutions to problems, rather than relying on a single optimal solution. This explorative behavior is often attributed to noise in neuronal dynamics. However, equipping a neuronal circuit with noise does not necessarily lead to explorative dynamics unless the noise is correlated within ensembles. This modeling study introduces configurable, locally coherent noise to create explorative behavior in a neuronal sequence-processing circuit, contributing to understanding the neuronal mechanisms underlying decision strategies and sequential memory recall.
PLOS COMPUTATIONAL BIOLOGY
(2023)
Article
Physics, Multidisciplinary
Sandra Nestler, Moritz Helias, Matthieu Gilson
Summary: The study demonstrates a simple biologically inspired feedforward neuronal model that utilizes higher-order temporal fluctuations for information extraction in time series classification. By training synaptic weights and a non-linear gain function, the study reveals how nonlinearity allows for the transfer of higher-order correlations and maximization of classification accuracy. The experimental results highlight the advantage of the biologically inspired architecture in utilizing the number of trainable parameters compared to classical machine learning approaches.
PHYSICAL REVIEW RESEARCH
(2023)
Article
Physics, Multidisciplinary
Kirsten Fischer, Alexandre Rene, Christian Keup, Moritz Layer, David Dahmen, Moritz Helias
Summary: Understanding the functional principles of information processing in deep neural networks with trained weights is a challenge. This study focuses on the mapping between probability distributions implemented by a deep feed-forward network, characterizing it as an iterated transformation of distributions. The analysis reveals that correlations up to second order play a key role in capturing information processing in internal layers of the network.
PHYSICAL REVIEW RESEARCH
(2022)
Article
Engineering, Electrical & Electronic
Anno C. Kurth, Johanna Senk, Dennis Terhorst, Justin Finnerty, Markus Diesmann
Summary: Simulating full-scale neuronal network models of the brain is challenging, but this study achieved shorter run times than the simulated biological time span by using a recent conventional compute node. Realtime performance is important for robotics and closed-loop applications, while sub-realtime performance is desirable for studying learning and development in the brain, which takes hours and days.
NEUROMORPHIC COMPUTING AND ENGINEERING
(2022)
Article
Physics, Multidisciplinary
Claudia Merger, Timo Reinartz, Stefan Wessel, Carsten Honerkamp, Andreas Schuppert, Moritz Helias
Summary: Networks with fat-tailed degree distributions often have hubs, nodes with high numbers of connections, crucial to the transition into a globally ordered network state. Higher order interaction effects counteract the self-feedback on hubs, highlighting their importance for the distinct onset of local versus global order in the network. This mechanism may be relevant for other systems with a strongly hierarchical underlying network structure.
PHYSICAL REVIEW RESEARCH
(2021)