Bulletin of the American Physical Society
APS March Meeting 2021
Monday–Friday, March 15–19, 2021; Virtual; Time Zone: Central Daylight Time, USA
Session F17: Stochastic Thermodynamics of Biological and Artificial Information Processing  IILive

Hide Abstracts 
Sponsoring Units: GSNP DCOMP Chair: Artemy Kolchinsky, Santa Fe Inst 
Tuesday, March 16, 2021 11:30AM  11:42AM Live 
F17.00001: Entropic cost to accurately solve the boolean satisfiability problem using a stochastic discrete system. Hadrien Vroylandt, Jonah Greenberg, Todd Gingrich It has long been recognized that information processing is intimately intertwined with the physical system that carries out that processing. This connection implies that information processing is associated with an entropic cost, as famously expressed in Landauer's bound for memory erasure. A more general example of information processing is with the satisfiability problem, whose difficulty can fall into different computational classes such as P or NP. Proposals have been put forwards to gain insights into NPproblems through their expressions as chemical networks. At the same time, computation in such systems is inherently stochastic, meaning the accuracy of the computation is an interesting quantity. I will present a set of results that connect the accuracy to the entropic cost of computation in stochastic processes. Based on explicit examples of stochastic systems aimed at solving the boolean satisfiability problem, we study the thermodynamics of the problem leading to an accuracycost tradeoff. This tradeoff is shown to be systemsize dependent. 
Tuesday, March 16, 2021 11:42AM  11:54AM Live 
F17.00002: Stochastic thermodynamics in uncertain environments Jan Korbel, David Wolpert Conventional stochastic thermodynamics assumes perfect knowledge of the parameters governing how a system interacts with its environment (number of baths, their temperatures, chemical potentials, precise rate matrices, etc.). In reality, none of these parameters is known precisely. We explore the thermodynamic implications of such uncertainty in the parameter vector. We assume a probability distribution over parameter vectors, which is sampled before the process begins, and the system then evolves according to that (unknown) parameter vector. We first investigate whether the modified definitions of the standard thermodynamic quantities such that the parameteraveraged state distribution evolving under the parameteraveraged parameter vector will obey the standard laws of thermodynamics. As we show, the first law is still obeyed with parameteraveraged definitions of heat, work, and internal energy. However, the Shannon entropy of the parameteraveraged state distribution can violate the second law. On the other hand, we can use the parameteraveraged stochastic (i.e., trajectorylevel) entropy to construct two quantities fulfilling the second law at the ensemble level. We investigate the relation between them and establish a connection to observable quantities in an experiment. 
Tuesday, March 16, 2021 11:54AM  12:06PM Live 
F17.00003: Thermodynamic Uncertainty Relations for Multipartite Processes Gülce Kardes, David Wolpert The thermodynamic uncertainty relations (TURs) provide upper bounds on the precision of an arbitrary current in a nonequilibrium system in terms of the entropy production of that system. All TURs derived so far have concerned single physical systems. However, many physical scenarios of interest involve multiple interacting systems, e.g. organelles within a biological cell. Here we show how to extend the previously derived TURs to those scenarios by exploiting a recently introduced framework of the thermodynamics of interacting systems [arXiv: 2003.11144]. These extended TURs bound the global EP over a set of interacting systems in terms of a weighted sum of the precisions of local currents emerging within those systems  plus an informationtheoretic correction term. This informationtheoretic correction term is a generalization of the multiinformation among the states of the overall system. We also obtain expressions concerning the relations between the local EPs of the individual systems and the statistical coupling among the currents within these systems. In particular, we derive such results for both the cases of scalarvalued and vectorvalued currents within each system. We illustrate our results with numerical experiments. 
Tuesday, March 16, 2021 12:06PM  12:18PM Live 
F17.00004: Information Phase Transitions in Random Classical Circuits Anasuya Lyons, Soonwon Choi, Ehud Altman Recent studies showed that the interplay between unitary dynamics and projective measurements leads to a novel phase transition characterized by the dynamics of quantum entanglement, entropy, and the Fisher information. In this work, we study a classical analogue of the phase transition by investigating random classical circuits interspersed by biterasure errors in one dimension. We find evidence of a phase transition from an “encoding” to a “nonencoding” phase above a critical error rate. In the encoding phase, a nonzero amount of Shannon entropy per bit can be retained in the system, indicating at least a subset of initially encoded information is protected from errors. In contrast, in the nonencoding phase, the Shannon entropy approaches zero within a finite circuit depth independent of system sizes. We extract the phase transition point and the dynamical scaling exponents near the transition point using numerical simulations. Furthermore, we develop an exact method to map the average behavior of the classical circuit dynamics to an infectious disease model, providing a new framework to study universal aspects in the dynamics of classical information. 
Tuesday, March 16, 2021 12:18PM  12:30PM Live 
F17.00005: A Stochastic Model for Logic Circuits Chloe Gao, Gavin E. Crooks, David Limmer We introduce a minimal stochastic model for complementary logic gates built with fieldeffect transistors. We characterise the performance of such gates with tools from information theory, and study the interplay between the accuracy, speed, and dissipation of computations. With a few universal building blocks, such as the NOT gate and the NAND gate, we are able to build arbitrary combinational and sequential logic circuits, and implement computing tasks. Our work provides a platform to study designing principles for lowdissipation digital devices, within the framework of stochastic thermodynamics. 
Tuesday, March 16, 2021 12:30PM  12:42PM 
F17.00006: FiniteTime Landauer Principle Karel Proesmans, Jannik Ehrich, John Bechhoefer Landauer's principle presents a lower bound on the energy dissipation needed to erase information. The bound, however, can only be reached when the erasure process is quasistatic. Here, we address the thermodynamic cost associated with the erasure of one bit of information over a finite amount of time. We present a general framework for minimizing the average work required when full control of a system’s microstates is possible. In addition to exact numerical results, we find simple bounds proportional to the variance of the microscopic distribution associated with the state of the bit. In the shorttime limit, we get a closed expression for the minimum average amount of work needed to erase a bit. The average work associated with the optimal protocol can be up to a factor of four smaller relative to protocols constrained to end in local equilibrium. Assessing prior experimental and numerical results based on heuristic protocols, we find that our bounds often dissipate an order of magnitude less energy. 
Tuesday, March 16, 2021 12:42PM  12:54PM Live 
F17.00007: Stochastic Thermodynamics of NonLinear Electronic Circuits: A Realistic Framework for Thermodynamics of Computation Jose Freitas, JeanCharles Delvenne, Massimiliano Esposito The rigorous description of intrinsic thermal noise in complex, nonlinear and out of equilibrium electronic circuits is a problem of fundamental as well as practical importance, of relevance for the design of new computing schemes that are energetically efficient. In this contribution we develop a formalism that allows us to construct thermodynamically consistent stochastic models of arbitrary circuits. It is thus able to accomodate large classes of technologically relevant circuits (for example, single electron and CMOS devices). The formalism is based on the theory of stochastic thermodynamics, which allows to provide a full thermodynamic characterization of the system and its entropy production, and to derive different fluctuation theorems. As a first application, we perform a full analysis of a CMOS inverter, or NOT gate. Based on this elementary example, we propose a design for a binary stochastic neuron, which employs intrinsic thermal noise as a resource. This can be considered as a generator of random bits with controllable rate and bias, and is the basis for the physical implementation of artificial neural networks or stochastic annealers. 
Tuesday, March 16, 2021 12:54PM  1:06PM Live 
F17.00008: The blearyeyed Maxwell demon: a mutual informationfueled colloidal engine Govind Paneru, Sandipan Dutta, Takahiro Sagawa, Tsvi Tlusty, Hyuk Kyu Pak The MaxwellmeetsShannon problem of a demon who observes a thermal system through a noisy information channel is omnipresent in nonequilibrium physics, and in particular in living systems where signaling and perception are often prone to noise. Yet, the direct measurement of the informationenergy interplay of these demons with bleary vision has so far been elusive for continuous systems. Here, by directly controlling and gauging the capacity of the demon’s detection channel, we extract the full mutual informationwork performance curve for a cyclic colloidal engine operating in nonequilibrium steadystate and study the efficiency fluctuations. We find that the most efficient engines, are far from perfect and utilize only about 0.51 bits of positional information per cycle. At this noise level of maximal average efficiency, the distribution of efficiency switches from bimodal to unimodal, and the stochastic efficiency often exceeds the bound set by the generalized second law. We identify a line of anomalous, noisedriven equilibrium states that defines a refrigeratortoheater transition. Crossing this line, the thermodynamic measurables and their fluctuations switch their behavior, and the generalized integral fluctuation theorem passes through unity. 
Tuesday, March 16, 2021 1:06PM  1:18PM Live 
F17.00009: The Energetic Cost of Different Biological Strategies to Transfer Information Samuel Bryant, Benjamin B Machta One of the primary computational requirements of a cellular system is the ability to transfer information between spatially separated components. In real systems, there is a wide variety of qualitatively different physical strategies for accomplishing this task, each suited to a different set of constraints. Electrical signaling is used by ion channels to communicate with voltage sensitive receptors. Diffusive signaling is used by kinases which produce second messengers which travel either through the membrane or bulk to reach a target. Order parameters of phase transitions are used to localize targets to specific regions. For each strategy, an energy consuming sender transmits a signal by modifying its local environment and the signal is broadcast to distant receivers. Here we explore the energetic cost of sending a signal as a function of frequency and spatial separation between the sender and receiver in the context of these strategies, computing a lower bound on the energetic cost of sending a bit of information. This tradeoff between energy and information, driven by the different physical constraints, may explain why cells use diverse strategies for different signaling purposes. 
Tuesday, March 16, 2021 1:18PM  1:30PM Live 
F17.00010: Density matrix formulation of dynamical systems Swetamber Das, Jason Green Some physical systems respond to perturbations with cascading failures and others respond by transducing flows of energy and entropy to form structures or do work. The statistical evolution of perturbations is critical to mitigating disasters and to the ability to function dynamically. Laws governing the statistical evolution of ensembles are central to classical mechanics, quantum mechanics, and dynamical systems, but do not explicitly describe the spread of perturbations. Here, we establish a density matrix formalism for this purpose that applies to classical dynamical systems and is analogous to the density matrix formulation of quantum mechanics. In this statisticaldynamical framework, the classical density matrix describes the statistical state of a system, with the time evolution of unit Lyapunov vectors in the tangent space giving unitary dynamics, and the classical Liouville equation corresponding to the preservation of a trace, a feature typically associated with quantization. We illustrate the theory with conservative and dissipative model dynamics. 
Tuesday, March 16, 2021 1:30PM  1:42PM Live 
F17.00011: Skewed Thermodynamic Geometry and Optimal Free Energy Estimation Steven Blaber, David Sivak Accurate and precise measurements of free energy differences are crucial to the determination of stable phases of matter and the design of novel pharmaceutical ligands for targeted protein binding. We extend the thermodynamicgeometry framework to higherorder moments of energy dissipation, accounting for timereversal asymmetry, which allows us to design protocols (dynamic variations of control parameters) that maximize the accuracy and precision of free energy estimators. In a model system, we find that these designed protocols can improve the precision by a factor of 34, and accuracy by over a factor of 10. 
Tuesday, March 16, 2021 1:42PM  1:54PM Live 
F17.00012: Thermodynamics of NonElementary Chemical Reaction Networks Francesco Avanzini, Gianmaria Falasco, Massimiliano Esposito We develop a thermodynamic framework for closed and open chemical networks applicable to nonelementary reactions that do not need to obey mass action kinetics. It only requires the knowledge of the kinetics and of the standard chemical potentials, and makes use of the topological properties of the network (conservation laws and cycles). Our approach is proven to be exact if the network results from a bigger network of elementary reactions where the fastevolving species have been coarse grained. Our work should be particularly relevant for energetic considerations in biosystems where the characterization of the elementary dynamics is seldomly achieved. 
Tuesday, March 16, 2021 1:54PM  2:06PM Live 
F17.00013: Irreversibility in biological active matter Junang Li, Jordan Horowitz, Nikta Fakhri Nonequilibrium dynamics is an essential physical feature of living matter. Living systems harness energy at the molecularscale through ATP hydrolysis and dissipate it on much larger spatiotemporal scales, often in the form of heat. The energetic loss can be cast as an increase in entropy of the environment, and the entropy production is associated with broken timereversal symmetry in the system’s dynamics. Here we estimate the entropy production rate by analyzing statistical properties of a time series observed in a nonequilibrium steady state. The KullbackLeibler divergence (KLD) between the time series and its time reversed is a lower bound to the entropy production rate. We use a lossless compression algorithm to quantify the information loss between forward and backward processes. The compression algorithm provides a universal measurement of irreversibility independent of the underlying models. 
Tuesday, March 16, 2021 2:06PM  2:18PM 
F17.00014: Information geometry of chemical thermodynamics Kohei Yoshimura, Sosuke Ito We study a connection between chemical thermodynamics and information geometry for the rate equation, where unnormalized concentration distributions are of importance rather than probability distributions. We introduce information geometry related to the Gibbs free energy of an ideal dilute solution, and discuss its thermodynamic interpretation. From a viewpoint of information geometry, we obtain a speed limit for a changing rate of the Gibbs free energy, a general bound of chemical fluctuations, and a tradeoff relation between speed and time. We also discuss its application to biochemical reaction. 
Tuesday, March 16, 2021 2:18PM  2:30PM Live 
F17.00015: Drivendissipative dynamics of active cytoskeletal networks underlie nearcritical energy fluctuations Carlos Floyd, Herbert Levine, Christopher Jarzynski, Garegin A. Papoian Eukaryotic cells are mechanically supported by a polymer network called the cytoskeleton, which consumes chemical energy to dynamically remodel its structure. Information processing occurs in cytoskeletal networks through the coupling of chemical signaling pathways to structuralmechanical responses which produce useful morphological changes. Recent experiments revealed that cytoskeletal remodeling occasionally happens through unusually large, steplike entropy production events reminiscent of earthquakes. These “cytoquakes" may reflect an optimization of information processing, and are hence of significant theoretical interest. The physics underlying cytoquakes is poorly understood, however, hindering investigation of their possible biological roles. Here we use agentbased simulations with a computational routine for quantifying entropy production to show that cytoquakes' origins lie in the inherent drivendissipative dynamics of active cytoskeletal networks, similarly to models exhibiting selforganized criticality. Combining machine learning with normal mode decomposition, we show that mechanical instability precedes cytoquakes, which then induce a spatial homogenization of tension sustained by the network. 
Follow Us 
Engage
Become an APS Member 
My APS
Renew Membership 
Information for 
About APSThe American Physical Society (APS) is a nonprofit membership organization working to advance the knowledge of physics. 
© 2021 American Physical Society
 All rights reserved  Terms of Use
 Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 207403844
(301) 2093200
Editorial Office
1 Research Road, Ridge, NY 119612701
(631) 5914000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 200452001
(202) 6628700