arXiv subjects

Peter Sollich

Publications and source records attributed to Peter Sollich.

At least 91 records · Page 5Linked to original sources

arXiv2017

Critical scaling in hidden state inference for linear Langevin dynamics

We consider the problem of inferring the dynamics of unknown (i.e. hidden) nodes from a set of observed trajectories and study analytically the average prediction error and the typical relaxation time of correlations between errors. We focus on a stochastic linear dynamics of continuous degrees of freedom interacting via random Gaussian couplings in the infinite network size limit. The expected error on the hidden time courses can be found as the equal-time hidden-to-hidden covariance of the probability distribution conditioned on observations. In the stationary regime, we analyze the phase diagram in the space of relevant parameters, namely the ratio between the numbers of observed and hidden nodes, the degree of symmetry of the interactions and the amplitudes of the hidden-to-hidden and hidden-to-observed couplings relative to the decay constant of the internal hidden dynamics. In particular, we identify critical regions in parameter space where the relaxation time and the inference error diverge, and determine the corresponding scaling behaviour.

cond-mat.dis-nn↗

arXiv2017

Inference for dynamics of continuous variables: the Extended Plefka Expansion with hidden nodes

We consider the problem of a subnetwork of observed nodes embedded into a larger bulk of unknown (i.e. hidden) nodes, where the aim is to infer these hidden states given information about the subnetwork dynamics. The biochemical networks underlying many cellular and metabolic processes are important realizations of such a scenario as typically one is interested in reconstructing the time evolution of unobserved chemical concentrations starting from the experimentally more accessible ones. We present an application to this problem of a novel dynamical mean field approximation, the Extended Plefka Expansion, which is based on a path integral description of the stochastic dynamics. As a paradigmatic model we study the stochastic linear dynamics of continuous degrees of freedom interacting via random Gaussian couplings. The resulting joint distribution is known to be Gaussian and this allows us to fully characterize the posterior statistics of the hidden nodes. In particular the equal-time hidden-to-hidden variance -- conditioned on observations -- gives the expected error at each node when the hidden time courses are predicted based on the observations. We assess the accuracy of the Extended Plefka Expansion in predicting these single node variances as well as error correlations over time, focussing on the role of the system size and the number of observed nodes.

cond-mat.stat-mech↗

arXiv2017

Statistics of non-affine defect precursors: tailoring defect densities in colloidal crystals using external fields

Coarse-graining atomic displacements in a solid produces both local affine strains and "non-affine" fluctuations. Here we study the equilibrium dynamics of these coarse grained quantities to obtain space-time dependent correlation functions. We show how a subset of these thermally excited, non-affine fluctuations act as precursors for the nucleation of lattice defects and suggest how defect probabilities may be altered by an {\it experimentally realisable} "external" field conjugate to the global non-affinity parameter. Our results are amenable to verification in experiments on colloidal crystals using commonly available holographic laser tweezer and video microscopy techniques, and may lead to simple ways of controlling the defect density of a colloidal solid.

cond-mat.soft↗

arXiv2017

Inferring hidden states in Langevin dynamics on large networks: Average case performance

We present average performance results for dynamical inference problems in large networks, where a set of nodes is hidden while the time trajectories of the others are observed. Examples of this scenario can occur in signal transduction and gene regulation networks. We focus on the linear stochastic dynamics of continuous variables interacting via random Gaussian couplings of generic symmetry. We analyze the inference error, given by the variance of the posterior distribution over hidden paths, in the thermodynamic limit and as a function of the system parameters and the ratio α between the number of hidden and observed nodes. By applying Kalman filter recursions we find that the posterior dynamics is governed by an "effective" drift that incorporates the effect of the observations. We present two approaches for characterizing the posterior variance that allow us to tackle, respectively, equilibrium and nonequilibrium dynamics. The first appeals to Random Matrix Theory and reveals average spectral properties of the inference error and typical posterior relaxation times, the second is based on dynamical functionals and yields the inference error as the solution of an algebraic equation.

cond-mat.dis-nn↗

arXiv2016

Amorphization of a crystalline solid by plastic deformation

We demonstrate that plastic deformation in solids is associated with a dynamic transition that is reminiscent to the transition from a superconducting to a mixed phase in Type II superconductors. We report analytic calculations, extensive molecular dynamics and sequential umbrella sampling Monte Carlo simulations of a two dimensional triangular crystalline solid undergoing plastic deformation under strain. The solid consists of particles connected by harmonic springs. Non-affine displacement fluctuations of the solid are enhanced using an external field, causing a rich deformation behaviour. The external field leads to a mixed phase, where defect and stress-free crystallites are surrounded by a network of amorphous regions with large local internal stress --- a "stress Meissner" effect. The transition shows slow ageing dynamics caused by the presence of many competing, non-crystalline free-energy minima. Under uniform uniaxial strain, this amorphization transition gives rise to irreversible plastic deformation with the amorphous regions appearing as band-like structures. Our results may be checked by careful experiments on colloidal crystals using holographic optical tweezers.

cond-mat.soft↗

arXiv2016

Equilibrium and dynamic pleating of a crystalline bonded network

We describe a phase transition that gives rise to structurally non-trivial states in a two-dimensional ordered network of particles connected by harmonic bonds. Monte Carlo simulations reveal that the network supports, apart from the homogeneous phase, a number of heterogeneous "pleated" phases, which can be stabilised by an external field. This field is conjugate to a global collective variable quantifying "non-affineness", i.e.~the deviation of local particle displacements from local affine deformation. In the pleated phase, stress is localised in ordered rows of pleats and eliminated from the rest of the lattice. The {\em kinetics} of the phase transition is unobservably slow in molecular dynamics simulation near coexistence, due to very large free energy barriers. When the external field is increased further to lower these barriers, the network exhibits rich dynamic behaviour: it transforms into a {\em metastable} phase with the stress now localised in a {\em disordered} arrangement of pleats. The pattern of pleats shows ageing dynamics and slow relaxation to equilibrium. Our predictions may be checked by experiments on tethered colloidal solids in dynamic laser traps.

cond-mat.soft↗

arXiv2016

Aging and linear response in the Hébraud-Lequeux model for amorphous rheology

We analyse the aging dynamics of the Hébraud-Lequeux model, a self-consistent stochastic model for the evolution of local stress in an amorphous material. We show that the model exhibits initial-condition dependent freezing: the stress diffusion constant decays with time as $D\sim 1/t^2$ during aging so that the cumulative amount of memory that can be erased, which is given by the time integral of $D(t)$, is finite. Accordingly the shear stress relaxation function, which we determine in the long-time regime, only decays to a plateau and becomes progressively elastic as the system ages. The frequency-dependent shear modulus exhibits a corresponding overall decay of the dissipative part with system age, while the characteristic relaxation times scale linearly with age as expected.

cond-mat.soft↗

arXiv2016

Stochastic evolution in populations of ideas

It is known that learning of players who interact in a repeated game can be interpreted as an evolutionary process in a population of ideas. These analogies have so far mostly been established in deterministic models, and memory loss in learning has been seen to act similarly to mutation in evolution. We here propose a representation of reinforcement learning as a stochastic process in finite "populations of ideas". The resulting birth-death dynamics has absorbing states and allows for the extinction or fixation of ideas, marking a key difference to mutation-selection processes in finite populations. We characterize the outcome of evolution in populations of ideas for several classes of symmetric and asymmetric games.

q-bio.PE↗

arXiv2016

Michaelis-Menten dynamics in protein subnetworks

To understand the behaviour of complex systems it is often necessary to use models that describe the dynamics of subnetworks. It has previously been established using projection methods that such subnetwork dynamics generically involves memory of the past, and that the memory functions can be calculated explicitly for biochemical reaction networks made up of unary and binary reactions. However, many established network models involve also Michaelis-Menten kinetics, to describe e.g. enzymatic reactions. We show that the projection approach to subnetwork dynamics can be extended to such networks, thus significantly broadening its range of applicability. To derive the extension we construct a larger network that represents enzymes and enzyme complexes explicitly, obtain the projected equations, and finally take the limit of fast enzyme reactions that gives back Michaelis-Menten kinetics. The crucial point is that this limit can be taken in closed form. The outcome is a simple procedure that allows one to obtain a description of subnetwork dynamics, including memory functions, starting directly from any given network of unary, binary and Michaelis-Menten reactions. Numerical tests show that this closed form enzyme elimination gives a much more accurate description of the subnetwork dynamics than the simpler method that represents enzymes explicitly, and is also more efficient computationally.

q-bio.MN↗

arXiv2016

Path integral methods for the dynamics of stochastic and disordered systems

We review some of the techniques used to study the dynamics of disordered systems subject to both quenched and fast (thermal) noise. Starting from the Martin-Siggia-Rose path integral formalism for a single variable stochastic dynamics, we provide a pedagogical survey of the perturbative, i.e. diagrammatic, approach to dynamics and how this formalism can be used for studying soft spin models. We review the supersymmetric formulation of the Langevin dynamics of these models and discuss the physical implications of the supersymmetry. We also describe the key steps involved in studying the disorder-averaged dynamics. Finally, we discuss the path integral approach for the case of hard Ising spins and review some recent developments in the dynamics of such kinetic Ising models.

cond-mat.dis-nn↗

arXiv2016

Extended Plefka Expansion for Stochastic Dynamics

We propose an extension of the Plefka expansion, which is well known for the dynamics of discrete spins, to stochastic differential equations with continuous degrees of freedom and exhibiting generic nonlinearities. The scenario is sufficiently general to allow application to e.g. biochemical networks involved in metabolism and regulation. The main feature of our approach is to constrain in the Plefka expansion not just first moments akin to magnetizations, but also second moments, specifically two-time correlations and responses for each degree of freedom. The end result is an effective equation of motion for each single degree of freedom, where couplings to other variables appear as a self-coupling to the past (i.e. memory term) and a coloured noise. This constitutes a new mean field approximation that should become exact in the thermodynamic limit of a large network, for suitably long-ranged couplings. For the analytically tractable case of linear dynamics we establish this exactness explicitly by appeal to spectral methods of Random Matrix Theory, for Gaussian couplings with arbitrary degree of symmetry.

cond-mat.dis-nn↗

arXiv2015

Elastic models of the glass transition applied to a liquid with density anomalies

Elastic models of the glass transition relate the relaxation dynamics and the elastic properties of structural glasses. They are based on the assumption that the relaxation dynamics occurs through activated events in the energy landscape whose energy scale is set by the elasticity of the material. Here we investigate whether such elastic models describe the relaxation dynamics of systems of particles interacting via a purely repulsive harmonic potential, focusing on a volume fraction and temperature range that is characterized by entropy--driven water--like density anomalies. We do find clear correlations between relaxation time and diffusivity on the one hand, and plateau shear modulus and Debye--Waller factor on the other, thus supporting the validity of elastic models of the glass transition. However, we also show that the plateau shear modulus is not related to the features of the underlying energy landscape of the system, at variance with recent results for power--law potentials. This challenges the common potential energy landscape interpretation of elastic models.

cond-mat.soft↗

arXiv2015

Dynamics and instantaneous normal modes in a liquid with density anomalies

We investigate the relation between the dynamical features of a supercooled liquid and those of its potential energy landscape, focusing on a model liquid with density anomalies. We consider, at fixed temperature, pairs of state points with different density but the same diffusion constant, and find that surprisingly they have identical dynamical features at all length and time scales. This is shown by the collapse of their mean square displacements and of their self--intermediate scattering functions at different wavevectors. We then investigate how the features of the energy landscape change with density, and establish that state points with equal diffusion constant have different landscapes. In particular, we find a correlation between the fraction of instantaneous normal modes connecting different energy minima and the diffusion constant, but unlike in other systems these two quantities are not in one--to--one correspondence with each other, showing that additional landscape features must be relevant in determining the diffusion constant.

cond-mat.soft↗

arXiv2015

Large-scale fluctuations of the largest Lyapunov exponent in diffusive systems

We present a general formalism for computing the largest Lyapunov exponent and its fluctuations in spatially extended systems described by diffusive fluctuating hydrodynamics, thus extending the concepts of dynamical system theory to a broad range of non-equilibrium systems. Our analytical results compare favourably with simulations of a lattice model of heat conduction. We further show how the computation of the Lyapunov exponent for the Symmetric Simple Exclusion Process relates to damage spreading and to a two-species pair annihilation process, for which our formalism yields new finite size results.

cond-mat.stat-mech↗

arXiv2015

Increased signaling entropy in cancer requires the scale-free property of protein interaction networks

One of the key characteristics of cancer cells is an increased phenotypic plasticity, driven by underlying genetic and epigenetic perturbations. However, at a systems-level it is unclear how these perturbations give rise to the observed increased plasticity. Elucidating such systems-level principles is key for an improved understanding of cancer. Recently, it has been shown that signaling entropy, an overall measure of signaling pathway promiscuity, and computable from integrating a sample's gene expression profile with a protein interaction network, correlates with phenotypic plasticity and is increased in cancer compared to normal tissue. Here we develop a computational framework for studying the effects of network perturbations on signaling entropy. We demonstrate that the increased signaling entropy of cancer is driven by two factors: (i) the scale-free (or near scale-free) topology of the interaction network, and (ii) a subtle positive correlation between differential gene expression and node connectivity. Indeed, we show that if protein interaction networks were random graphs, described by Poisson degree distributions, that cancer would generally not exhibit an increased signaling entropy. In summary, this work exposes a deep connection between cancer, signaling entropy and interaction network topology.

q-bio.MN↗

arXiv2015

Speech Recognition Front End Without Information Loss

Speech representation and modelling in high-dimensional spaces of acoustic waveforms, or a linear transformation thereof, is investigated with the aim of improving the robustness of automatic speech recognition to additive noise. The motivation behind this approach is twofold: (i) the information in acoustic waveforms that is usually removed in the process of extracting low-dimensional features might aid robust recognition by virtue of structured redundancy analogous to channel coding, (ii) linear feature domains allow for exact noise adaptation, as opposed to representations that involve non-linear processing which makes noise adaptation challenging. Thus, we develop a generative framework for phoneme modelling in high-dimensional linear feature domains, and use it in phoneme classification and recognition tasks. Results show that classification and recognition in this framework perform better than analogous PLP and MFCC classifiers below 18 dB SNR. A combination of the high-dimensional and MFCC features at the likelihood level performs uniformly better than either of the individual representations across all noise levels.

cs.CL↗

arXiv2015

The average number of distinct sites visited by a random walker on random graphs

We study the linear large $n$ behavior of the average number of distinct sites $S(n)$ visited by a random walker after $n$ steps on a large random graph. An expression for the graph topology dependent prefactor $B$ in $S(n) = Bn$ is proposed. We use generating function techniques to relate this prefactor to the graph adjacency matrix and then devise message-passing equations to calculate its value. Numerical simulations are performed to evaluate the agreement between the message passing predictions and random walk simulations on random graphs. Scaling with system size and average graph connectivity are also analysed.

cond-mat.stat-mech↗

arXiv2015

Modeling the emergence of polarity patterns for the intercellular transport of auxin in plants

The hormone auxin is actively transported throughout plants via protein machineries including the dedicated transporter known as PIN. The associated transport is ordered with nearby cells driving auxin flux in similar directions. Here we provide a model of both the auxin transport and of the dynamics of cellular polarisation based on flux sensing. Our main findings are: (i) spontaneous intracellular PIN polarisation arises if PIN recycling dynamics are sufficiently non-linear, (ii) there is no need for an auxin concentration gradient, and (iii) ordered multi-cellular patterns of PIN polarisation are favored by molecular noise.

q-bio.TO↗