arXiv subjects

Saurav Talukdar

Publications and source records attributed to Saurav Talukdar.

At least 19 recordsLinked to original sources

arXiv2023

Hindering loads prompt clustered configurations that enhance stability during cargo transport by multiple Kinesin-1

Transport of intracellular cargo is often mediated by teams of molecular motors that function in a chaotic environment under varying conditions. We show that the motors have unique steady state behavior which enables transport modalities that are robust. Under reduced ATP concentrations, multi-motor configurations are preferred over single motors. Higher load force drives motors to cluster, but very high loads compel them to separate in a manner that promotes immediate cargo movement once the load subsides. These inferences, backed by analytical guarantees, provide unique insights into the coordination strategies adopted by molecular motors to transport intracellular cargo.

q-bio.SC↗

arXiv2022

Efficient and passive learning of networked dynamical systems driven by non-white exogenous inputs

We consider a networked linear dynamical system with $p$ agents/nodes. We study the problem of learning the underlying graph of interactions/dependencies from observations of the nodal trajectories over a time-interval $T$. We present a regularized non-casual consistent estimator for this problem and analyze its sample complexity over two regimes: (a) where the interval $T$ consists of $n$ i.i.d. observation windows of length $T/n$ (restart and record), and (b) where $T$ is one continuous observation window (consecutive). Using the theory of $M$-estimators, we show that the estimator recovers the underlying interactions, in either regime, in a time-interval that is logarithmic in the system size $p$. To the best of our knowledge, this is the first work to analyze the sample complexity of learning linear dynamical systems \emph{driven by unobserved not-white wide-sense stationary (WSS) inputs}.

cs.LG↗

arXiv2021

Power Modeling for Effective Datacenter Planning and Compute Management

Datacenter power demand has been continuously growing and is the key driver of its cost. An accurate mapping of compute resources (CPU, RAM, etc.) and hardware types (servers, accelerators, etc.) to power consumption has emerged as a critical requirement for major Web and cloud service providers. With the global growth in datacenter capacity and associated power consumption, such models are essential for important decisions around datacenter design and operation. In this paper, we discuss two classes of statistical power models designed and validated to be accurate, simple, interpretable and applicable to all hardware configurations and workloads across hyperscale datacenters of Google fleet. To the best of our knowledge, this is the largest scale power modeling study of this kind, in both the scope of diverse datacenter planning and real-time management use cases, as well as the variety of hardware configurations and workload types used for modeling and validation. We demonstrate that the proposed statistical modeling techniques, while simple and scalable, predict power with less than 5% Mean Absolute Percent Error (MAPE) for more than 95% diverse Power Distribution Units (more than 2000) using only 4 features. This performance matches the reported accuracy of the previous started-of-the-art methods, while using significantly less features and covering a wider range of use cases.

cs.DC↗

arXiv2021

Carbon-Aware Computing for Datacenters

The amount of CO$_2$ emitted per kilowatt-hour on an electricity grid varies by time of day and substantially varies by location due to the types of generation. Networked collections of warehouse scale computers, sometimes called Hyperscale Computing, emit more carbon than needed if operated without regard to these variations in carbon intensity. This paper introduces Google's system for Carbon-Intelligent Compute Management, which actively minimizes electricity-based carbon footprint and power infrastructure costs by delaying temporally flexible workloads. The core component of the system is a suite of analytical pipelines used to gather the next day's carbon intensity forecasts, train day-ahead demand prediction models, and use risk-aware optimization to generate the next day's carbon-aware Virtual Capacity Curves (VCCs) for all datacenter clusters across Google's fleet. VCCs impose hourly limits on resources available to temporally flexible workloads while preserving overall daily capacity, enabling all such workloads to complete within a day. Data from operation shows that VCCs effectively limit hourly capacity when the grid's energy supply mix is carbon intensive and delay the execution of temporally flexible workloads to "greener" times.

cs.DC↗

arXiv2021

Energetics of Feedback: Application to Memory Erasure

Landauer's erasure principle states that any irreversible erasure protocol of a single bit memory needs work of at least $k_B T ln2.$ Recent proof of concept experiments has demonstrated that the erasure protocols with work close to the Landauer limit can be devised. Under feedback, where the state of the bit can be measured, the work needed for bit erasure can be lower than $k_B T ln2.$ In this article, we analyze the energetics of feedback enabled erasure, while incorporating the imperfections of experimentally realized memory and bit erasure protocols that admit failure probabilities. We delineate the role of uncertainty in measurements and its effects on the work and entropy changes for a feedback-based erasure. We quantitatively demonstrate that the deficit between the Landauer limit and the minimum average work needed in a feedback-based erasure is accounted for by the mutual information between the measurement and the state of the memory, while incorporating the imperfections inherent in any realization. We experimentally demonstrate analysis results on a memory and erasure protocol realized using optical fields.

cs.IT↗

arXiv2020

Topology Learning in Radial Dynamical Systems with Unreliable Data

Many complex engineering systems admit bidirectional and linear couplings between their agents. Blind and passive methods to identify such influence pathways/couplings from data are central to many applications. However, dynamically related data-streams originating at different sources are prone to corruption caused by asynchronous time-stamps of different streams, packet drops and noise. Such imperfect information may be present in the entire observation period, and hence not detected by change-detection algorithms that require an initial clean observation period. Prior work has shown that spurious links are inferred in the graph structure due to the corrupted data-streams, which prevents consistent learning. In this article, we provide a novel approach to detect the location of corrupt agents as well as present an algorithm to learn the structure of radial dynamical systems despite corrupted data streams. In particular, we show that our approach provably learns the true radial structure if the unknown corrupted nodes are at least three hops away from each other. Our theoretical results are further validated in test dynamical network.

eess.SY↗

arXiv2020

The Differential Entropy of Mixtures: New Bounds and Applications

Mixture distributions are extensively used as a modeling tool in diverse areas from machine learning to communications engineering to physics, and obtaining bounds on the entropy of probability distributions is of fundamental importance in many of these applications. This article provides sharp bounds on the entropy concavity deficit, which is the difference between the entropy of the mixture and the weighted sum of entropies of constituent components. Toward establishing lower and upper bounds on the concavity deficit, results that are of importance in their own right are obtained. In order to obtain nontrivial upper bounds, properties of the skew-divergence are developed and notions of "skew" $f$-divergences are introduced; a reverse Pinsker inequality and a bound on Jensen-Shannon divergence are obtained along the way. Complementary lower bounds are derived with special attention paid to the case that corresponds to independent summation of a continuous and a discrete random variable. Several applications of the bounds are delineated, including to mutual information of additive noise channels, thermodynamics of computation, and functional inequalities.

cs.IT↗

arXiv2019

Graphical Models in Meshed Distribution Grids: Topology estimation, change detection and limitations

Graphical models are a succinct way to represent the structure in probability distributions. This article analyzes the graphical model of nodal voltages in non-radial power distribution grids. Using algebraic and structural properties of graphical models, algorithms exactly determining topology and detecting line changes for distribution grids are presented along with their theoretical limitations. We show that if distribution grids have cycles/loops of size greater than three, then nodal voltages are sufficient for efficient topology estimation without additional assumptions on system parameters. In contrast, line failure or change detection using nodal voltages does not require any structural assumption. Under noisy measurements, we provide the first non-trivial bounds on the maximum noise that the system can tolerate for asymptotically correct topology recovery. The performance of the designed algorithms is validated with nonlinear AC power flow samples generated by Matpower on test grids, including scenarios with injection correlations and system noise.

eess.SY↗

arXiv2019

Exact Topology Learning in a Network of Cyclostationary Processes

Learning the structure of a network from time series data, in particular cyclostationary data, is of significant interest in many disciplines such as power grids, biology and finance. In this article, an algorithm is presented for reconstruction of the topology of a network of cyclostationary processes. To the best of our knowledge, this is the first work to guarantee exact recovery without any assumptions on the underlying structure. The method is based on a lifting technique by which cyclostationary processes are mapped to vector wide sense stationary processes and further on semi-definite properties of matrix Wiener filters for the said processes.We demonstrate the performance of the proposed algorithm on a Resistor-Capacitor network and present the accuracy of reconstruction for varying sample sizes.

eess.SY↗

arXiv2018

Data-driven identification of a thermal network in multi-zone building

System identification of smart buildings is necessary for their optimal control and application in demand response. The thermal response of a building around an operating point can be modeled using a network of interconnected resistors with capacitors at each node/zone called RC network. The development of the RC network involves two phases: obtaining the network topology, and estimating thermal resistances and capacitance's. In this article, we present a provable method to reconstruct the interaction topology of thermal zones of a building solely from temperature measurements. We demonstrate that our learning algorithm accurately reconstructs the interaction topology for a $5$ zone office building in EnergyPlus with real-world conditions. We show that our learning algorithm is able to recover the network structure in scenarios where prior research prove insufficient.

eess.SY↗

arXiv2018

Physics Informed Topology Learning in Networks of Linear Dynamical Systems

Learning influence pathways of a network of dynamically related processes from observations is of considerable importance in many disciplines. In this article, influence networks of agents which interact dynamically via linear dependencies are considered. An algorithm for the reconstruction of the topology of interaction based on multivariate Wiener filtering is analyzed. It is shown that for a vast and important class of interactions, that respect flow conservation, the topology of the interactions can be exactly recovered. The class of problems where reconstruction is guaranteed to be exact includes power distribution networks, dynamic thermal networks and consensus networks. The efficacy of the approach is illustrated through simulation and experiments on consensus networks, IEEE power distribution networks and thermal dynamics of buildings.

eess.SY↗

arXiv2018

Realizing Information Erasure in Finite Time

In this article, we focus on erasure of a bit of information in finite time. Landauer's principle states that the average heat dissipation due to erasure of information is k_B T ln 2, which is achievable only in an asymptotic manner. Recent theoretical developments in non-equilibrium thermodynamics and stochastic control, predict a more general bound for finite time erasure dependent on the Wasserstein distances between the initial and final configurations. These predictions suggest improvements to experimental protocol with regards to minimizing average heat dissipation for bit erasure in finite time from a bistable well, under overdamped Langevin dynamics. We present a comparative study of a theoretically optimal protocol with an existing protocol, and highlight the closeness and deviation from optimality

cond-mat.stat-mech↗

arXiv2018

Error Bounds on a Mixed Entropy Inequality

Motivated by the entropy computations relevant to the evaluation of decrease in entropy in bit reset operations, the authors investigate the deficit in an entropic inequality involving two independent random variables, one continuous and the other discrete. In the case where the continuous random variable is Gaussian, we derive strong quantitative bounds on the deficit in the inequality. More explicitly it is shown that the decay of the deficit is sub-Gaussian with respect to the reciprocal of the standard deviation of the Gaussian variable. What is more, up to rational terms these results are shown to be sharp.

cs.IT↗

arXiv2018

Designing Memory Bits with Dissipation lower than the Landauer's Bound

A Brownian particle in a symmetric double well potential is used as a representation for a single bit memory, where, the location of the particle in either well denotes one of the two states of a single bit memory. This article analyzes the effect of modifications to a symmetric double well potential on the minimum heat dissipation associated with erasure of the information stored in a single bit memory. Two types of modifications are considered, viz., overlap between the two wells and the asymmetry between the two wells of a bit of memory. Moreover, the analysis presented here, takes into account the uncertainty in the success of the erasure process. We quantify the effect of the proposed modifications on the heat dissipation accompanying erasure of a bit of information with a comparison to the Landauer's bound. In particular, we conclude that the proposed modifications could result in the minimum heat dissipation being lower than the Landauer's bound in quasi-static erasure processes. Furthermore, we quantify the physical aspect ratio for designing a memory bit which could improve memory density.

cond-mat.stat-mech↗

arXiv2018

Topology Learning of Radial Dynamical Systems with Latent Nodes

In this article, we present a method to reconstruct the topology of a partially observed radial network of linear dynamical systems with bi-directional interactions. Our approach exploits the structure of the inverse power spectral density matrix and recovers edges involving nodes up to four hops away in the underlying topology. We then present an algorithm with provable guarantees, which eliminates the spurious links obtained and also identifies the location of the unobserved nodes in the inferred topology. The algorithm recovers the exact topology of the network by using only time-series of the states at the observed nodes. The effectiveness of the method developed is demonstrated by applying it on a typical distribution system of the electric grid.

eess.SY↗

arXiv2017

Learning the Exact Topology of Undirected Consensus Networks

In this article, we present a method to learn the interaction topology of a network of agents undergoing linear consensus updates in a non invasive manner. Our approach is based on multivariate Wiener filtering, which is known to recover spurious edges apart from the true edges in the topology. The main contribution of this work is to show that in the case of undirected consensus networks, all spurious links obtained using Wiener filtering can be identified using frequency response of the Wiener filters. Thus, the exact interaction topology of the agents is unveiled. The method presented requires time series measurements of the state of the agents and does not require any knowledge of link weights. To the best of our knowledge this is the first approach that provably reconstructs the structure of undirected consensus networks with correlated noise. We illustrate the effectiveness of the method developed through numerical simulations as well as experiments on a five node network of Raspberry Pis.

eess.SY↗

arXiv2017

Single Molecule Studies Under Constant Force Using Model Based Robust Control Design

Optical tweezers have enabled important insights into intracellular transport through the investigation of motor proteins, with their ability to manipulate particles at the microscale, affording femto Newton force resolution. Its use to realize a constant force clamp has enabled vital insights into the behavior of motor proteins under different load conditions. However, the varying nature of disturbances and the effect of thermal noise pose key challenges to force regulation. Furthermore, often the main aim of many studies is to determine the motion of the motor and the statistics related to the motion, which can be at odds with the force regulation objective. In this article, we propose a mixed objective H2-Hinfinity optimization framework using a model-based design, that achieves the dual goals of force regulation and real time motion estimation with quantifiable guarantees. Here, we minimize the Hinfinity norm for the force regulation and error in step estimation while maintaining the H2 norm of the noise on step estimate within user specified bounds. We demonstrate the efficacy of the framework through extensive simulations and an experimental implementation using an optical tweezer setup with live samples of the motor protein kinesin; where regulation of forces below 1 pico Newton with errors below 10 percent is obtained while simultaneously providing real time estimates of motor motion.

math.OC↗

arXiv2017

Distributed Apportioning in a Power Network for providing Demand Response Services

Greater penetration of Distributed Energy Resources (DERs) in power networks requires coordination strategies that allow for self-adjustment of contributions in a network of DERs, owing to variability in generation and demand. In this article, a distributed scheme is proposed that enables a DER in a network to arrive at viable power reference commands that satisfies the DERs local constraints on its generation and loads it has to service, while, the aggregated behavior of multiple DERs in the network and their respective loads meet the ancillary services demanded by the grid. The Net-load Management system for a single unit is referred to as the Local Inverter System (LIS) in this article . A distinguishing feature of the proposed consensus based solution is the distributed finite time termination of the algorithm that allows each LIS unit in the network to determine power reference commands in the presence of communication delays in a distributed manner. The proposed scheme allows prioritization of Renewable Energy Sources (RES) in the network and also enables auto-adjustment of contributions from LIS units with lower priority resources (non-RES). The methods are validated using hardware-in-the-loop simulations with Raspberry PI devices as distributed control units, implementing the proposed distributed algorithm and responsible for determining and dispatching realtime power reference commands to simulated power electronics interface emulating LIS units for demand response.

eess.SY↗