-
Signatures of polarized chiral spin disproportionation in rare earth nickelates
Authors:
Jiarui Li,
Robert J. Green,
Claribel Domínguez,
Abraham Levitan,
Yi Tseng,
Sara Catalano,
Jennifer Fowlie,
Ronny Sutarto,
Fanny Rodolakis,
Lucas Korol,
Jessica L. McChesney,
John W. Freeland,
Dirk Van der Marel,
Marta Gibert,
Riccardo Comin
Abstract:
In rare earth nickelates (RENiO$_3$), electron-lattice coupling drives a concurrent metal-to-insulator and bond disproportionation phase transition whose microscopic origin has long been the subject of active debate. Of several proposed mechanisms, here we test the hypothesis that pairs of self-doped ligand holes spatially condense to provide local spin moments that are antiferromagnetically coupl…
▽ More
In rare earth nickelates (RENiO$_3$), electron-lattice coupling drives a concurrent metal-to-insulator and bond disproportionation phase transition whose microscopic origin has long been the subject of active debate. Of several proposed mechanisms, here we test the hypothesis that pairs of self-doped ligand holes spatially condense to provide local spin moments that are antiferromagnetically coupled to Ni spins. These singlet-like states provide a basis for long-range bond and spiral spin order. Using magnetic resonant X-ray scattering on NdNiO$_3$ thin films, we observe the chiral nature of the spin-disproportionated state, with spin spirals propagating along the crystallographic (101)$_\mathrm{ortho}$ direction. These spin spirals are found to preferentially couple to X-ray helicity, establishing the presence of a hitherto-unobserved macroscopic chirality. The presence of this chiral magnetic configuration suggests a potential multiferroic coupling between the noncollinear magnetic arrangement and improper ferroelectric behavior as observed in prior studies on NdNiO$_3$ (101)$_\mathrm{ortho}$ films and RENiO$_3$ single crystals. Experimentally constrained theoretical double-cluster calculations confirm the presence of an energetically stable spin-disproportionated state with Zhang-Rice singlet-like combinations of Ni and ligand moments.
△ Less
Submitted 5 September, 2024; v1 submitted 30 August, 2024;
originally announced August 2024.
-
An active metasurface enhanced with moiré ferroelectricity
Authors:
Dong Seob Kim,
Chengxin Xiao,
Roy C. Dominguez,
Zhida Liu,
Hamza Abudayyeh,
Kyoungpyo Lee,
Rigo Mayorga-Luna,
Hyunsue Kim,
Kenji Watanabe,
Takashi Taniguchi,
Chih-Kang Shih,
Yoichi Miyahara,
Wang Yao,
Xiaoqin Li
Abstract:
Semiconductor moiré systems, characterized by their periodic spatial light emission, unveil a new paradigm of active metasurfaces. Here, we show that ferroelectric moiré domains formed in a twisted hexagonal boron nitride (t-hBN) substrate can modulate light emission from an adjacent semiconductor MoSe$_2$ monolayer, enhancing its functionality as an active metasurface. The electrostatic potential…
▽ More
Semiconductor moiré systems, characterized by their periodic spatial light emission, unveil a new paradigm of active metasurfaces. Here, we show that ferroelectric moiré domains formed in a twisted hexagonal boron nitride (t-hBN) substrate can modulate light emission from an adjacent semiconductor MoSe$_2$ monolayer, enhancing its functionality as an active metasurface. The electrostatic potential at the surface of the t-hBN substrate provides a simple way to confine excitons in the MoSe$_2$ monolayer. The excitons confined within the domains and at the domain walls are spectrally separated due to a pronounced Stark shift. Moreover, the patterned light emission can be dynamically controlled by electrically gating the ferroelectric domains, introducing a novel functionality beyond conventional semiconductor moiré systems. Our findings chart an exciting pathway for integrating nanometer-scale moiré ferroelectric domains with various optically active functional layers, paving the way for advanced nanophotonic applications.
△ Less
Submitted 17 August, 2024; v1 submitted 17 May, 2024;
originally announced May 2024.
-
Emergent electronic landscapes in a novel valence-ordered nickelate with tri-component nickel coordination
Authors:
Aravind Raji,
Zhengang Dong,
Victor Porée,
Alaska Subedi,
Xiaoyan Li,
Bernat Mundet,
Lucia Varbaro,
Claribel Domínguez,
Marios Hadjimichael,
Bohan Feng,
Alessandro Nicolaou,
Jean-Pascal Rueff,
Danfeng Li,
Alexandre Gloter
Abstract:
The metal-hydride-based topochemical reduction process has produced novel thermodynamically unstable phases across various transition metal oxide series with unusual crystal structures and non-trivial ground states. Here, by such an oxygen (de-) intercalation method we synthesis a novel samarium nickelate with ordered nickel valences associated with tri-component coordination configurations. This…
▽ More
The metal-hydride-based topochemical reduction process has produced novel thermodynamically unstable phases across various transition metal oxide series with unusual crystal structures and non-trivial ground states. Here, by such an oxygen (de-) intercalation method we synthesis a novel samarium nickelate with ordered nickel valences associated with tri-component coordination configurations. This structure, with a formula of Sm$_{9}$Ni$_{9}$O$_{22}$ as revealed by four-dimensional scanning transmission electron microscopy, emerges from the intricate planes of {303}$_{\text{pc}}$ ordered apical oxygen vacancies. X-ray spectroscopy measurements and ab-initio calculations show the coexistence of square-planar, pyramidal and octahedral Ni sites with mono-, bi- and tri-valences. It leads to an intense orbital polarization, charge-ordering, and a ground state with a strong electron localization marked by the disappearance of ligand-hole configuration at low-temperature. This new nickelate compound provides another example of previously inaccessible materials enabled by topotactic transformations and presents a unique platform where mixed Ni valence can give rise to exotic phenomena.
△ Less
Submitted 5 August, 2023;
originally announced August 2023.
-
Mapping Orthorhombic Domains with Geometrical Phase Analysis in Rare-Earth Nickelate Heterostructures
Authors:
Bernat Mundet,
Marios Hadjimichael,
Jennifer Fowlie,
Lukas Korosec,
Lucia Varbaro,
Claribel Dominguez,
Jean-Marc Triscone,
Duncan T. L. Alexander
Abstract:
Most perovskite oxides belong to the Pbnm space group, composed by an anisotropic unit cell, A-site antipolar displacements and oxygen octahedral tilts. Mapping the orientation of the orthorhombic unit cell in epitaxial heterostructures that consist of at least one Pbnm compound is often required to understand and control the different degrees of coupling established at their coherent interfaces a…
▽ More
Most perovskite oxides belong to the Pbnm space group, composed by an anisotropic unit cell, A-site antipolar displacements and oxygen octahedral tilts. Mapping the orientation of the orthorhombic unit cell in epitaxial heterostructures that consist of at least one Pbnm compound is often required to understand and control the different degrees of coupling established at their coherent interfaces and, therefore, their resulting physical properties. However, retrieving this information from the strain maps generated with high-resolution scanning transmission electron microscopy can be challenging, because the three pseudocubic lattice parameters are very similar in these systems. Here, we present a novel methodology for mapping the crystallographic orientation in Pbnm systems. It makes use of the geometrical phase analysis algorithm, as applied to aberration-corrected scanning transition electron microscopy images, but in an unconventional way. The method is fast and robust, giving real-space maps of the lattice orientations in Pbnm systems, from both cross-sectional and plan-view geometries and across large fields of view. As an example, we apply our methodology to rare-earth nickelate heterostructures, in order to investigate how the crystallographic orientation of these films depends on various structural constraints that are imposed by the underlying single crystal substrates. We observe that the resulting domain distributions and associated defect landscapes mainly depend on a competition between the epitaxial compressive/tensile and shear strains, together with the matching of atomic displacements at the substrate/film interface. The results point towards strategies for controlling these characteristics by appropriate substrate choice.
△ Less
Submitted 13 July, 2023;
originally announced July 2023.
-
Electrostatic moiré potential from twisted-hBN layers
Authors:
Dong Seob Kim,
Roy C. Dominguez,
Rigo Mayorga-Luna,
Dingyi Ye,
Jacob Embley,
Tixuan Tan,
Yue Ni,
Zhida Liu,
Mitchell Ford,
Frank Y. Gao,
Saba Arash,
Kenji Watanabe,
Takashi Taniguchi,
Suenne Kim,
Chih-Kang Shih,
Keji Lai,
Wang Yao,
Li Yang,
Xiaoqin Li,
Yoichi Miyahara
Abstract:
Moiré superlattices formed by vertically stacking van der Waals layers host a rich variety of correlated electronic phases and function as novel photonic materials. The moiré potential of the superlattice, however, is fixed by the interlayer coupling of the stacked functional layers (e.g. graphene) and dependent on carrier types (e.g. electrons or holes) and valleys (e.g. Γ vs. K). In contrast, tw…
▽ More
Moiré superlattices formed by vertically stacking van der Waals layers host a rich variety of correlated electronic phases and function as novel photonic materials. The moiré potential of the superlattice, however, is fixed by the interlayer coupling of the stacked functional layers (e.g. graphene) and dependent on carrier types (e.g. electrons or holes) and valleys (e.g. Γ vs. K). In contrast, twisted hexagonal boron nitride (hBN) layers are predicted to impose a periodic electrostatic potential that may be used to engineer the properties of an adjacent functional thin layer. Here, we show that this potential is described by a simple theory of electric polarization originating from the interfacial charge redistribution, validated by its dependence on supercell sizes and distance from the twisted interfaces. We demonstrate that the potential depth and profile can be further controlled by assembling a double moiré structure. When the twist angles are similar at the two interfaces, the potential is deepened by adding the potential from the two twisted interfaces, reaching ~ 400 meV. When the twist angles are dissimilar at the two interfaces, multi-level polarization states are observed. As an example of controlling a functional layer, we demonstrate how the electrostatic potential from a twisted hBN substrate impedes exciton diffusion in a semiconductor monolayer. These findings suggest exciting opportunities for engineering properties of an adjacent functional layer using the surface potential of a twisted hBN substrate.
△ Less
Submitted 13 June, 2023;
originally announced June 2023.
-
Characteristic lengthscales of the electrically-induced insulator-to-metal transition
Authors:
Theodor Luibrand,
Adrien Bercher,
Rodolfo Rocco,
Farnaz Tahouni-Bonab,
Lucia Varbaro,
Carl Willem Rischau,
Claribel Domínguez,
Yixi Zhou,
Weiwei Luo,
Soumen Bag,
Lorenzo Fratino,
Reinhold Kleiner,
Stefano Gariglio,
Dieter Koelle,
Jean-Marc Triscone,
Marcelo J. Rozenberg,
Alexey B. Kuzmenko,
Stefan Guénon,
Javier del Valle
Abstract:
Some correlated materials display an insulator-to-metal transition as the temperature is increased. In most cases this transition can also be induced electrically, resulting in volatile resistive switching due to the formation of a conducting filament. While this phenomenon has attracted much attention due to potential applications, many fundamental questions remain unaddressed. One of them is its…
▽ More
Some correlated materials display an insulator-to-metal transition as the temperature is increased. In most cases this transition can also be induced electrically, resulting in volatile resistive switching due to the formation of a conducting filament. While this phenomenon has attracted much attention due to potential applications, many fundamental questions remain unaddressed. One of them is its characteristic lengths: what sets the size of these filaments, and how does this impact resistive switching properties. Here we use a combination of wide-field and scattering-type scanning near-field optical microscopies to characterize filament formation in NdNiO3 and SmNiO3 thin films. We find a clear trend: smaller filaments increase the current density, yielding sharper switching and a larger resistive drop. With the aid of numerical simulations, we discuss the parameters controlling the filament width and, hence, the switching properties.
△ Less
Submitted 1 January, 2023;
originally announced January 2023.
-
Coupling of Magnetic Phases at Nickelate Interfaces
Authors:
C. Domínguez,
J. Fowlie,
A. B. Georgescu,
B. Mundet,
N. Jaouen,
M. Viret,
A. Suter,
A. J. Millis,
Z. Salman,
T. Prokscha,
M. Gibert,
J. -M. Triscone
Abstract:
In this work we present a model system built out of artificially layered materials, allowing us to understand the interrelation of magnetic phases with that of the metallic-insulating phase at long length-scales, and enabling new strategies for the design and control of materials in devices. The artificial model system consists of superlattices made of SmNiO$_3$ and NdNiO$_3$ layers -- two members…
▽ More
In this work we present a model system built out of artificially layered materials, allowing us to understand the interrelation of magnetic phases with that of the metallic-insulating phase at long length-scales, and enabling new strategies for the design and control of materials in devices. The artificial model system consists of superlattices made of SmNiO$_3$ and NdNiO$_3$ layers -- two members of the fascinating rare earth nickelate family, having different metal-to-insulator and magnetic transition temperatures. By combining two complementary techniques -- resonant elastic x-ray scattering and muon spin relaxation -- we show how the magnetic order evolves, in this complex multicomponent system, as a function of temperature and superlattice periodicity. We demonstrate that the length scale of the coupling between the antiferromagnetic and paramagnetic phases is longer than that of the electronic metal-insulator phase transition -- despite being subsidiary to it. This can be explained via a Landau theory -- where the bulk magnetic energy plus a gradient cost between magnetic and non magnetic phases are considered. These results provide a clear understanding of the coupling of magnetic transitions in systems sharing identical order parameters.
△ Less
Submitted 12 November, 2022;
originally announced November 2022.
-
Robust ferromagnetism in insulating La$_2$NiMnO$_6$ thin films
Authors:
G. De Luca,
J. Spring,
U. Bashir,
M. Campanini,
R. Totani,
C. Dominguez,
A. Zakharova,
M. Döbeli,
T. Greber,
M. D. Rossell,
C. Piamonteze,
M. Gibert
Abstract:
The field of oxide spintronics can strongly benefit from the establishment of robust ferromagnetic insulators with near room-temperature Curie temperature. Here we investigate the structural, electronic, and magnetic properties of atomically-precise epitaxially-strained thin films of the double perovskite La$_2$NiMnO$_6$ (LNMO) grown by off-axis radio-frequency magnetron sputtering. We find that t…
▽ More
The field of oxide spintronics can strongly benefit from the establishment of robust ferromagnetic insulators with near room-temperature Curie temperature. Here we investigate the structural, electronic, and magnetic properties of atomically-precise epitaxially-strained thin films of the double perovskite La$_2$NiMnO$_6$ (LNMO) grown by off-axis radio-frequency magnetron sputtering. We find that the films retain both a strong insulating behavior and a bulk-like Curie temperature in the order of 280 K, nearly independently from epitaxial strain conditions. These results suggest a prospective implementation of LNMO films in multi-layer device architectures where a high-temperature ferromagnetic insulating state is a prerequisite.
△ Less
Submitted 20 January, 2021; v1 submitted 19 January, 2021;
originally announced January 2021.
-
Probing photo-induced rearrangements in the NdNiO$_{3}$ magnetic spiral with polarization-sensitive ultrafast resonant soft x-ray scattering
Authors:
K. R. Beyerlein,
A. S. Disa,
M Först,
M. Henstridge,
T. Gebert,
T. Forrest,
A. Fitzpatrick,
C. Dominguez,
J. Fowlie,
M. Gibert,
J. -M. Triscone,
S. S. Dhesi,
A. Cavalleri
Abstract:
We use resonant soft X-ray diffraction to track the photo-induced dynamics of the antiferromagnetic structure in a NdNiO$_{3}$ thin film. Femtosecond laser pulses with a photon energy of 0.61 eV, resonant with electron transfer between long-bond and short-bond nickel sites, are used to excite the material and drive an ultrafast insulator-metal transition. Polarization sensitive soft X-ray diffract…
▽ More
We use resonant soft X-ray diffraction to track the photo-induced dynamics of the antiferromagnetic structure in a NdNiO$_{3}$ thin film. Femtosecond laser pulses with a photon energy of 0.61 eV, resonant with electron transfer between long-bond and short-bond nickel sites, are used to excite the material and drive an ultrafast insulator-metal transition. Polarization sensitive soft X-ray diffraction, resonant to the nickel L$_{3}$-edge, then probes the evolution of the underlying magnetic spiral as a function of time delay with 80 picosecond time resolution. By modelling the azimuthal dependence of the scattered intensity for different linear X-ray polarizations, we benchmark the changes of the local magnetic moments and the spin alignment. The measured changes are consistent with a reduction of the long-bond site magnetic moments and an alignment of the spins towards a more collinear structure at early time delays.
△ Less
Submitted 18 June, 2020; v1 submitted 14 April, 2020;
originally announced April 2020.
-
Length-scales of interfacial coupling between metal-insulator phases in oxides
Authors:
Claribel Domínguez,
Alexandru B. Georgescu,
Bernat Mundet,
Yajun Zhang,
Jennifer Fowlie,
Alain Mercy,
Sara Catalano,
Duncan T. L. Alexander,
Philippe Ghosez,
Antoine Georges,
Andrew J. Millis,
Marta Gibert,
Jean-Marc Triscone
Abstract:
Controlling phase transitions in transition metal oxides remains a central feature of both technological and fundamental scientific relevance. A well-known example is the metal-insulator transition which has been shown to be highly controllable while a less well understood aspect of this phenomenon is the length scale over which the phases can be established. To gain further insight into this issu…
▽ More
Controlling phase transitions in transition metal oxides remains a central feature of both technological and fundamental scientific relevance. A well-known example is the metal-insulator transition which has been shown to be highly controllable while a less well understood aspect of this phenomenon is the length scale over which the phases can be established. To gain further insight into this issue, we have atomically engineered an artificially phase separated system through fabricating epitaxial superlattices consisting of SmNiO$_{3}$ and NdNiO$_{3}$, two materials undergoing a metal-to-insulator transition at different temperatures. By combining advanced experimental techniques and theoretical modeling, we demonstrate that the length scale of the metal-insulator transition is controlled by the balance of the energy cost of the domain wall between a metal and insulator and the bulk energetics. Notably, we show that the length scale of this effect exceeds that of the physical coupling of structural motifs, introducing a new paradigm for interface-engineering properties that are not available in bulk
△ Less
Submitted 17 October, 2019; v1 submitted 15 October, 2019;
originally announced October 2019.
-
Effect of surfactant concentration on the responsiveness of a thermoresponsive copolymer/surfactant mixture with potential application on Smart foams formulations
Authors:
M. M. Soledad Lencina,
Eugenio Fernández Miconi,
Marcos D. Fernández Leyes,
Claudia Domínguez,
Ezequiel Cuenca,
Hernán A. Ritacco
Abstract:
We studied a system formed by a mixture of a thermoresponsive negatively charged graft copolymer (Alg-g-PNIPAAm) with a brush-type structure, and an oppositely charged surfactant (DTAB), in bulk and at the air-solution interface. We performed experiments of surface tension, electrophoretic mobility, dynamic and static light scattering and atomic force microscopy in order to characterize the comple…
▽ More
We studied a system formed by a mixture of a thermoresponsive negatively charged graft copolymer (Alg-g-PNIPAAm) with a brush-type structure, and an oppositely charged surfactant (DTAB), in bulk and at the air-solution interface. We performed experiments of surface tension, electrophoretic mobility, dynamic and static light scattering and atomic force microscopy in order to characterize the complexes formed as a function of DTAB concentration and temperature. We found that these polymer-surfactant complexes are able to respond by changing their sizes, both in bulk and at the air-solution interface, when T is increased above the coil-globule transition temperature (LSCT) of the copolymer. However, the thermoresponse was found to be dependent on surfactant concentration, cs: for cs < 2.8 mM, the size of the aggregates decreases as T increases but, for cs >= 2.8 mM, the opposite behavior takes place, i.e. the size increases with T. At the interface, the intensity of the effect produced on the surface tension by increasing T above LCST diminishes continuously as cs increases, reducing the ability of the interfacial complex to respond to temperature changes. We studied the stability of aqueous foams formulated with these mixtures as a function of T and cs. We found that the stability of the foam can be modulated by changing T, but we observed that this effect is dependent on the surfactant concentration range. We found a correlation between changes in the aggregates sizes, the surface tension behavior and the responsiveness of foam stability to changes of temperature.
△ Less
Submitted 8 July, 2017;
originally announced July 2017.
-
Time evolution of the extremely diluted Blume-Emery-Griffiths neural network
Authors:
D. Bolle',
D. R. C. Dominguez,
R. Erichsen Jr.,
E. Korutcheva,
W. K. Theumann
Abstract:
The time evolution of the extremely diluted Blume-Emery-Griffiths neural network model is studied, and a detailed equilibrium phase diagram is obtained exhibiting pattern retrieval, fluctuation retrieval and self-sustained activity phases. It is shown that saddle-point solutions associated with fluctuation overlaps slow down considerably the flow of the network states towards the retrieval fixed…
▽ More
The time evolution of the extremely diluted Blume-Emery-Griffiths neural network model is studied, and a detailed equilibrium phase diagram is obtained exhibiting pattern retrieval, fluctuation retrieval and self-sustained activity phases. It is shown that saddle-point solutions associated with fluctuation overlaps slow down considerably the flow of the network states towards the retrieval fixed points. A comparison of the performance with other three-state networks is also presented.
△ Less
Submitted 14 August, 2002; v1 submitted 14 August, 2002;
originally announced August 2002.
-
Numerical simulation of a binary communication channel: Comparison between a replica calculation and an exact solution
Authors:
D. R. C. Dominguez,
M. Maravall,
A. Turiel,
J. C. Ciria,
N. Parga
Abstract:
The mutual information of a single-layer perceptron with $N$ Gaussian inputs and $P$ deterministic binary outputs is studied by numerical simulations. The relevant parameters of the problem are the ratio between the number of output and input units, $α= P/N$, and those describing the two-point correlations between inputs. The main motivation of this work refers to the comparison between the repl…
▽ More
The mutual information of a single-layer perceptron with $N$ Gaussian inputs and $P$ deterministic binary outputs is studied by numerical simulations. The relevant parameters of the problem are the ratio between the number of output and input units, $α= P/N$, and those describing the two-point correlations between inputs. The main motivation of this work refers to the comparison between the replica computation of the mutual information and an analytical solution valid up to $α\sim O(1)$. The most relevant results are: (1) the simulation supports the validity of the analytical prediction, and (2) it also verifies a previously proposed conjecture that the replica solution interpolates well between large and small values of $α$.
△ Less
Submitted 14 July, 2001;
originally announced July 2001.
-
Categorization in fully connected multi-state neural network models
Authors:
R. Erichsen,
W. K. Theumann,
D. R. C. Dominguez
Abstract:
The categorization ability of fully connected neural network models, with either discrete or continuous Q-state units, is studied in this work in replica symmetric mean-field theory. Hierarchically correlated multi-state patterns in a two level structure of ancestors and descendents (examples) are embedded in the network and the categorization task consists in recognizing the ancestors when the…
▽ More
The categorization ability of fully connected neural network models, with either discrete or continuous Q-state units, is studied in this work in replica symmetric mean-field theory. Hierarchically correlated multi-state patterns in a two level structure of ancestors and descendents (examples) are embedded in the network and the categorization task consists in recognizing the ancestors when the network is trained exclusively with their descendents. Explicit results for the dependence of the equilibrium properties of a Q=3-state model and a $Q=\infty$-state model are obtained in the form of phase diagrams and categorization curves. A strong improvement of the categorization ability is found when the network is trained with examples of low activity. The categorization ability is found to be robust to finite threshold and synaptic noise. The Almeida-Thouless lines that limit the validity of the replica-symmetric results, are also obtained.
△ Less
Submitted 1 October, 1999;
originally announced October 1999.
-
Information Capacity of a Hierarchical Neural Network
Authors:
David Renato Carreta Dominguez
Abstract:
The information conveyed by a hierarchical attractor neural network is examined. The network learns sets of correlated patterns (the examples) in the lowest level of the hierarchical tree and can categorize them at the upper levels. A way to measure the non-extensive information content of the examples is formulated. Curves showing the transition from a large retrieval information to a large cat…
▽ More
The information conveyed by a hierarchical attractor neural network is examined. The network learns sets of correlated patterns (the examples) in the lowest level of the hierarchical tree and can categorize them at the upper levels. A way to measure the non-extensive information content of the examples is formulated. Curves showing the transition from a large retrieval information to a large categorization information behavior, when the number of examples increase, are displayed. The conditions for the maximal information are given as functions of the correlation between examples and the load of concepts. Numerical simulations support the analytical results.
△ Less
Submitted 7 September, 1998;
originally announced September 1998.
-
Mutual Information of Three-State Low Activity Diluted Neural Networks with Self-Control
Authors:
D. Bolle',
D. R. C. Dominguez,
S. Amari
Abstract:
The influence of a macroscopic time-dependent threshold on the retrieval process of three-state extremely diluted neural networks is examined. If the threshold is chosen appropriately in function of the noise and the pattern activity of the network, adapting itself in the course of the time evolution, it guarantees an autonomous functioning of the network. It is found that this self-control mech…
▽ More
The influence of a macroscopic time-dependent threshold on the retrieval process of three-state extremely diluted neural networks is examined. If the threshold is chosen appropriately in function of the noise and the pattern activity of the network, adapting itself in the course of the time evolution, it guarantees an autonomous functioning of the network. It is found that this self-control mechanism considerably improves the retrieval quality, especially in the limit of low activity, including the storage capacity, the basins of attraction and the information content. The mutual information is shown to be the relevant parameter to study the retrieval quality of such low activity models. Numerical results confirm these observations.
△ Less
Submitted 21 August, 2000; v1 submitted 5 June, 1998;
originally announced June 1998.
-
Inference and Chaos by a Network of Non-monotonic Neurons
Authors:
D. R. C. Dominguez
Abstract:
The generalization properties of an attractive network of non monotonic neurons which infers concepts from samples are studied. The macroscopic dynamics for the overlap between the state of the neurons with the concepts, well as the activity of the neurons, are obtained and searched for through its numerical behavior. Complex behavior leading from fixed points to chaos through a cascade of bifur…
▽ More
The generalization properties of an attractive network of non monotonic neurons which infers concepts from samples are studied. The macroscopic dynamics for the overlap between the state of the neurons with the concepts, well as the activity of the neurons, are obtained and searched for through its numerical behavior. Complex behavior leading from fixed points to chaos through a cascade of bifurcation are found, when we increase the correlation between samples or decrease the activity of the samples and the load of concepts, or tune the threshold of fatigue of the neurons. Both the information dimension and the Liapunov exponent are given, and a phase diagram is built.
△ Less
Submitted 29 January, 1998;
originally announced January 1998.
-
Self-control in Sparsely Coded Networks
Authors:
D. R. C. Dominguez,
D. Bolle
Abstract:
A complete self-control mechanism is proposed in the dynamics of neural networks through the introduction of a time-dependent threshold, determined in function of both the noise and the pattern activity in the network. Especially for sparsely coded models this mechanism is shown to considerably improve the storage capacity, the basins of attraction and the mutual information content of the netwo…
▽ More
A complete self-control mechanism is proposed in the dynamics of neural networks through the introduction of a time-dependent threshold, determined in function of both the noise and the pattern activity in the network. Especially for sparsely coded models this mechanism is shown to considerably improve the storage capacity, the basins of attraction and the mutual information content of the network.
△ Less
Submitted 26 January, 1998;
originally announced January 1998.
-
Categorization by a three-state attractor neural network
Authors:
D. R. C. Dominguez,
D. Bollé
Abstract:
The categorization properties of an attractor network of three-state neurons which infers three-state concepts from examples are studied. The evolution equations governing the parallel dynamics at zero temperature for the overlap between the state of the network and the examples, the state of the network and the concepts as well as the neuron activity are discussed in the limit of extreme diluti…
▽ More
The categorization properties of an attractor network of three-state neurons which infers three-state concepts from examples are studied. The evolution equations governing the parallel dynamics at zero temperature for the overlap between the state of the network and the examples, the state of the network and the concepts as well as the neuron activity are discussed in the limit of extreme dilution. A transition from a retrieval region to a categorization region is found when the number of examples or their correlations are increased. If the pattern activity is small enough, the examples (concepts) are very well retrieved (categorized) for an appropriate choice of the zero-activity threshold of the neurons.
△ Less
Submitted 18 April, 1997;
originally announced April 1997.
