Alianna J. Maren - Academia.edu (original) (raw)

Papers by Alianna J. Maren

88ER12824, entitled "Enhancing Nuclear Power Plant Performance through the Use of Artificial Inte... more 88ER12824, entitled "Enhancing Nuclear Power Plant Performance through the Use of Artificial Intel i igence." Enclosed is the first annual report for the period September 30, 1988 to May 31, 1989 on the above referenced contract• It consists of an "Overview" of the progress made in the various projects, along with a several supporting internal reports that describe the work in considerable detail. In several cases, papers have been published or submitted for publication are referenced. Copies of these reports are also attached. Also, two papers describing what may be "patentable ideas" were mubmltted on June i, 1989 to "-DOE patent counsel in the DOE Chicago Operations Office for patent review. These reports

The objective of this research is to advance the state-of-the-art of applying artificial intellig... more The objective of this research is to advance the state-of-the-art of applying artificial intelligence technology (both expert systems and neural networks) to enhancing the performance (safety, efficiency, control and management) of nuclear power plants. A second, but equally important, objective is to build a broadly based critical mass of expertise in the artificial intelligence field that can be brought to bear on the technology of nuclear power plants. This means the production of graduates at the B.S., M.S., and Ph.D. levels in Nuclear Engineering and related fields. The research undertaken for this program is particularly appropriate for the M.S. theses and Ph.D. dissertations. A third objective is to transfer the technology developed to the nuclear power community,'' as well as the scientific and technological community,'' through publications in appropriate journals and proceedings and through presentations at national and international meetings.

Journal of Solid State Chemistry, Jul 1, 1984

Hysteresis in the pressure-dependent solid state phase transition Pr,012-PrgOl, is modeled using ... more Hysteresis in the pressure-dependent solid state phase transition Pr,012-PrgOl, is modeled using a thermodynamic formalism. The system is considered to be formed of a fixed number of domains, which are differentiated on the basis of size. The two cases of noninteracting and interacting domains are considered. The interacting domains model allows a better fit to experimental results. In each case. the model is applied to four different isothermal hysteresis curves for the Pr70,2-Pr90ih phase transition. The kinetics of the phase transition are studied for the case of noninteracting domains.

arXiv (Cornell University), Sep 8, 2022

One of the biggest challenges in characterizing 2-D topographies is succinctly communicating the ... more One of the biggest challenges in characterizing 2-D topographies is succinctly communicating the dominant nature of local configurations. In a 2-D grid composed of bistate units, this could be expressed as finding the characteristic configuration variables such as nearestneighbor pairs and triplet combinations. The 2-D cluster variation method (CVM) provides a theoretical framework for associating a set of configuration variables with only two parameters, for a system that is at free energy equilibrium. This work presents a method for determining which of many possible two-parameter sets provides the "most suitable" match for a given 2-D topography, drawing from methods used for variational inference. This particular work focuses exclusively on topographies for which the activation enthalpy parameter (ε 0) is zero, so that the distribution between two states is equiprobable. This condition is used since, when the two states are equiprobable, there is an analytic solution giving the configuration variable values as functions of the h-value, where we define h in terms of the interaction enthalpy parameter (ε 1) as h = exp(2ε 1). This allows the computationally-achieved configuration variable values to be compared with the analytically-predicted values for a given h-value. The method is illustrated using four patterns derived from three different naturally-occurring black-and-white topographies, where each pattern meets the equiprobability criterion. We achieve expected results, that is, as the patterns progress from having relatively low numbers of like-near-like nodes to increasing like-near-like masses, the h-values for each corresponding free energyminimized model also increase. Further, the corresponding configuration variable values for the (free energy-minimized) model patterns are in close alignment with the analytically-predicted values. The method described here has applicability beyond characterizing specific 2-D topographies. Potential applications extend to active inference as well as to 2-D CORTECONs, which incorporate free energy minimization into a 2-D grid of latent variables, in addition to the usual methods employed with energy-based neural networks.

Sensor Fusion II, 1989

A major problem with MultiSensor Information Fusion (MSIF) is establishing the level of processin... more A major problem with MultiSensor Information Fusion (MSIF) is establishing the level of processing at which information should be fused. Current methodologies, whether based on fusion at the data element, segment/feature, or symbolic levels, are each inadequate for robust MSIF. Data-element fusion has problems with coregistration. Attempts to fuse information using the features of segmented data relies on a presumed similarity between the segmentation characteristics of each data stream. Symbolic-level fusion requires too much advance processing (including object identification) to be useful. MSIF systems need to operate in real-time, must perform fusion using a variety of sensor types, and should be effective across a wide range of operating conditions or deployment environments. We address this problem through developing a new representation level which facilitates matching and information fusion. The Hierarchical Data Structure (HDS) representation, created using a multilayer, cooperative/competitive neural network, meets this need. The HDS is an intermediate representation between the raw or smoothed data stream and symbolic interpretation of the data. It represents the structural organization of the data. Fused HDSs will incorporate information from multiple sensors. Their knowledge-rich structure aids top-down scene interpretation via both model matching and knowledge-based region interpretation.

ACM SIGART Bulletin, 1988

[Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics

The authors report on a prototype implementation and preliminary studies of a new class of comput... more The authors report on a prototype implementation and preliminary studies of a new class of computational engine. This engine introduces statistical mechanical considerations into a simple neural network design, affording greater stability in the pattern classes generated in response to different input stimulus. The current instantiation of the engine consists of two 1-D layers, with feedforward connections between the input

The derivation of key equations for the variational Bayes approach is well-known in certain circl... more The derivation of key equations for the variational Bayes approach is well-known in certain circles. However, translating the fundamental derivations (e.g., as found in Beal (2003)) to the notation of Friston (2013, 2015) is somewhat delicate. Further, the notion of using variational Bayes in the context of a system with Markov blankets requires special attention. This Technical Report presents the derivation in detail. It further illustrates how the variational Bayes method provides a framework for a new computational engine, incorporating the 2-D cluster variation method (CVM), which provides a necessary free energy equation that can be minimized across both the external and representational systems' states, respectively.

A new approach for general artificial intelligence (GAI), building on neural network deep learnin... more A new approach for general artificial intelligence (GAI), building on neural network deep learning architectures, can make use of one or more hidden layers that have the ability to continuously reach a free energy minimum even after input stimulus is removed, allowing for a variety of possible behaviors. One reason that this approach has not been developed until now has been the lack of a suitable free energy equation. The Cluster Variation Method (CVM) offers a means for characterizing 2-D local pattern distributions, or configuration variables, and provides a free energy formalism in terms of these configuration variables. The equilibrium distribution of these configuration variables is defined in terms of a single interaction enthalpy parameter, h, for the case of equiprobable distribution of bistate units. For non-equiprobable distributions, the equilibrium distribution can be characterized by providing a fixed value for the fraction of units in the active state (x1), correspond...

Part 1 Introduction - structures, dynamics, and learning: introduction, A.J.Maren history, C.T.Ha... more Part 1 Introduction - structures, dynamics, and learning: introduction, A.J.Maren history, C.T.Harston biological basis, C.T.Harston structures, A.J.Maren dynamics, A.J.Maren learning, C.T.Harston. Part 2 Neural network systems: multilayer, feedforward networks - delta learning rule, A.J.Maren multilayer, feedforward networks - optimizing learning rule, A.J.Maren single-layer, laterally connected networks, A.J.Maren single-layer, topologically organised networks, A.J.Maren bilayer, feedforward/feedback networks, A.J.Maren multilayer, co-operative/competitive networks, A.J.Maren systems of interacting networks, A.J.Maren. Part 3 Implementing neural networks: system design, D.Jones and S.Franklin configuring and optimizing feedforward networks, A.J.Maren et al hardware implementations, S.Morgan and C.R.Parten. Part 4 Neural network applications - group A - pattern recognition applications: spatio-temporal pattern recognition, A.J.Maren medical diagnosis, D.Jones sonar, P.Simpson radar...

SPIE Proceedings, 2015

A new information measure, drawing on the 1-D Cluster Variation Method (CVM), describes local pat... more A new information measure, drawing on the 1-D Cluster Variation Method (CVM), describes local pattern distributions (nearest-neighbor and next-nearest neighbor) in a binary 1-D vector in terms of a single interaction enthalpy parameter h for the specific case where the fractions of elements in each of two states are the same (x 1 =x 2 =0.5). An example application of this method would be for EEG interpretation in Brain-Computer Interfaces (BCIs), especially in the frontier of invariant biometrics based on distinctive and invariant individual responses to stimuli containing an image of a person with whom there is a strong affiliative response (e.g., to a person's grandmother). This measure is obtained by mapping EEG observed configuration variables (z 1 , z 2 , z 3 for next-nearest neighbor triplets) to h using the analytic function giving h in terms of these variables at equilibrium. This mapping results in a small phase space region of resulting h values, which characterizes local pattern distributions in the source data. The 1-D vector with equal fractions of units in each of the two states can be obtained using the method for transforming natural images into a binarized equiprobability ensemble (Saremi & Sejnowski, 2014; Stephens et al., 2013). An intrinsically 2-D data configuration can be mapped to 1-D using the 1-D Peano-Hilbert space-filling curve, which has demonstrated a 20 dB lower baseline using the method compared with other approaches (cf. SPIE ICA etc. by Hsu & Szu, 2014). This CVM-based method has multiple potential applications; one near-term one is optimizing classification of the EEG signals from a COTS 1-D BCI baseball hat. This can result in a convenient 3-D lab-tethered EEG, configured in a 1-D CVM equiprobable binary vector, and potentially useful for Smartphone wireless display. Longer-range applications include interpreting neural assembly activations via high-density implanted soft, cellular-scale electrodes.

Biological Cybernetics, 1993

We consider a model of a neuron coupled with a surrounding dendritic network subject to Langevin ... more We consider a model of a neuron coupled with a surrounding dendritic network subject to Langevin noise and a weak periodic modulation. Through an adiabatic elimination procedure, the single-neuron dynamics are extracted from the coupled stochastic differential equations describing the network of dendrodendritic interactions.Our approach yields a“reduced neuron” model whose dynamics may correspond to neurophysiologically realistic behavior for certain ranges

ArXiv, 2019

ArXiv, 2018

L'invention concerne un systeme pour effectuer une generation d'hypothese. Un processeur ... more L'invention concerne un systeme pour effectuer une generation d'hypothese. Un processeur d'extraction extrait une entite provenant d'un ensemble de donnees. Un processeur d'association associe l'entite extraite a un ensemble d'entites de reference pour obtenir une association potentielle, l'association potentielle entre l'unite extraite et l'ensemble d'entites de reference etant decrite a l'aide d'un ensemble de valeurs de croyances fondees sur des vecteurs. Un processeur de seuil determine si un ensemble de valeurs de croyances de l'ensemble de valeurs de croyances fondees sur des vecteurs depasse un seuil predetermine. Si les valeurs de croyances depassent un seuil predetermine, le processeur de seuil adopte cette association.

L'invention concerne un systeme d'architecture predictif et un procede qui supporte les e... more L'invention concerne un systeme d'architecture predictif et un procede qui supporte les evolutions futures previsionnelles grâce a l'agregation et a l'analyse de grandes quantites de donnees. Le systeme de l'invention comprend une architecture a multi-niveaux, presentant une ou plusieurs fonctions de retroaction qui commande la recherche et la decouverte d'informations dans le cas d'une « hypothese » ou la decouverte d'informations significatives avant la formation d'une hypothese claire et le passage sur un support predictif a niveau eleve ou un manque de support permettant l'existence et/ou une nouvelle hypothese.