Showing 1–43 of 43 results for author: Balakrishnan, R

Enhancing O-RAN Security: Evasion Attacks and Robust Defenses for Graph Reinforcement Learning-based Connection Management

Authors: Ravikumar Balakrishnan, Marius Arvinte, Nageen Himayat, Hosein Nikopour, Hassnaa Moustafa

Abstract: Adversarial machine learning, focused on studying various attacks and defenses on machine learning (ML) models, is rapidly gaining importance as ML is increasingly being adopted for optimizing wireless systems such as Open Radio Access Networks (O-RAN). A comprehensive modeling of the security threats and the demonstration of adversarial attacks and defenses on practical AI based O-RAN systems is… ▽ More Adversarial machine learning, focused on studying various attacks and defenses on machine learning (ML) models, is rapidly gaining importance as ML is increasingly being adopted for optimizing wireless systems such as Open Radio Access Networks (O-RAN). A comprehensive modeling of the security threats and the demonstration of adversarial attacks and defenses on practical AI based O-RAN systems is still in its nascent stages. We begin by conducting threat modeling to pinpoint attack surfaces in O-RAN using an ML-based Connection management application (xApp) as an example. The xApp uses a Graph Neural Network trained using Deep Reinforcement Learning and achieves on average 54% improvement in the coverage rate measured as the 5th percentile user data rates. We then formulate and demonstrate evasion attacks that degrade the coverage rates by as much as 50% through injecting bounded noise at different threat surfaces including the open wireless medium itself. Crucially, we also compare and contrast the effectiveness of such attacks on the ML-based xApp and a non-ML based heuristic. We finally develop and demonstrate robust training-based defenses against the challenging physical/jamming-based attacks and show a 15% improvement in the coverage rates when compared to employing no defense over a range of noise budgets △ Less

Submitted 6 May, 2024; originally announced May 2024.

Comments: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

Conditionally unutilized proteins and their profound effects on growth and adaptation across microbial species

Authors: Rohan Balakrishnan, Jonas Cremer

Abstract: Protein synthesis is an important determinant of microbial growth and response that demands a high amount of metabolic and biosynthetic resources. Despite these costs, microbial species from different taxa and habitats massively synthesize proteins that are not utilized in the conditions they currently experience. Based on resource allocation models, recent studies have begun to reconcile the cost… ▽ More Protein synthesis is an important determinant of microbial growth and response that demands a high amount of metabolic and biosynthetic resources. Despite these costs, microbial species from different taxa and habitats massively synthesize proteins that are not utilized in the conditions they currently experience. Based on resource allocation models, recent studies have begun to reconcile the costs and benefits of these conditionally unutilized proteins (CUPs) in the context of varying environmental conditions. Such massive synthesis of CUPs is crucial to consider in different areas of modern microbiology, from the systematic investigation of cell physiology, via the prediction of evolution in laboratory and natural environments, to the rational design of strains in biotechnology applications. △ Less

Submitted 19 June, 2023; originally announced June 2023.

Comments: 12 pages, 3 figures

Twisted curve geometry underlying topological invariants

Authors: Radha Balakrishnan, Rossen Dandoloff, Avadh Saxena

Abstract: Topological invariants such as winding numbers and linking numbers appear as charges of topological solitons in diverse nonlinear physical systems described by a unit vector field defined on two and three dimensional manifolds. While the Gauss-Bonnet theorem shows that the Euler characteristic (a topological invariant) can be written as the integral of the Gaussian curvature (an intrinsic geometri… ▽ More Topological invariants such as winding numbers and linking numbers appear as charges of topological solitons in diverse nonlinear physical systems described by a unit vector field defined on two and three dimensional manifolds. While the Gauss-Bonnet theorem shows that the Euler characteristic (a topological invariant) can be written as the integral of the Gaussian curvature (an intrinsic geometric quantity), the intriguing question of whether winding and linking numbers can also be expressed similarly as integrals of some intrinsic geometric quantities has not been addressed in the literature. In this paper we provide the answer by showing that for the winding number in two dimensions, these quantities are torsions of the two evolving space curves describing the manifold. On the other hand, in three dimensions we find that in addition to torsions, intrinsic twists of the space curves are necessary to obtain a nontrivial winding number and linking number. These new results arise from the hitherto unknown connections that we establish between these topological invariants and the corresponding appropriately normalized global anholonomies (i.e., geometric phases) associated with the unit vector fields on the respective manifolds. An application of our results to a 3D Heisenberg ferromagnetic model supporting a topological soliton is also presented. △ Less

Submitted 19 January, 2024; v1 submitted 12 April, 2023; originally announced April 2023.

Comments: 23 pages, 1 figure, 1 table

Journal ref: Phys. Lett. A 493, 129261 (2024)

Multi-Task Model Personalization for Federated Supervised SVM in Heterogeneous Networks

Authors: Aleksei Ponomarenko-Timofeev, Olga Galinina, Ravikumar Balakrishnan, Nageen Himayat, Sergey Andreev, Yevgeni Koucheryavy

Abstract: Federated systems enable collaborative training on highly heterogeneous data through model personalization, which can be facilitated by employing multi-task learning algorithms. However, significant variation in device computing capabilities may result in substantial degradation in the convergence rate of training. To accelerate the learning procedure for diverse participants in a multi-task feder… ▽ More Federated systems enable collaborative training on highly heterogeneous data through model personalization, which can be facilitated by employing multi-task learning algorithms. However, significant variation in device computing capabilities may result in substantial degradation in the convergence rate of training. To accelerate the learning procedure for diverse participants in a multi-task federated setting, more efficient and robust methods need to be developed. In this paper, we design an efficient iterative distributed method based on the alternating direction method of multipliers (ADMM) for support vector machines (SVMs), which tackles federated classification and regression. The proposed method utilizes efficient computations and model exchange in a network of heterogeneous nodes and allows personalization of the learning model in the presence of non-i.i.d. data. To further enhance privacy, we introduce a random mask procedure that helps avoid data inversion. Finally, we analyze the impact of the proposed privacy mechanisms and participant hardware and data heterogeneity on the system performance. △ Less

Submitted 1 April, 2023; v1 submitted 17 March, 2023; originally announced March 2023.

Comments: 14 pages, 12 figures, 4 tables, 1 algorithm; Added algorithm for iterative solution, updated the abstract, fixed typos

Stargazer: An Interactive Camera Robot for Capturing How-To Videos Based on Subtle Instructor Cues

Authors: Jiannan Li, Mauricio Sousa, Karthik Mahadevan, Bryan Wang, Paula Akemi Aoyaui, Nicole Yu, Angela Yang, Ravin Balakrishnan, Anthony Tang, Tovi Grossman

Abstract: Live and pre-recorded video tutorials are an effective means for teaching physical skills such as cooking or prototy** electronics. A dedicated cameraperson following an instructor's activities can improve production quality. However, instructors who do not have access to a cameraperson's help often have to work within the constraints of static cameras. We present Stargazer, a novel approach for… ▽ More Live and pre-recorded video tutorials are an effective means for teaching physical skills such as cooking or prototy** electronics. A dedicated cameraperson following an instructor's activities can improve production quality. However, instructors who do not have access to a cameraperson's help often have to work within the constraints of static cameras. We present Stargazer, a novel approach for assisting with tutorial content creation with a camera robot that autonomously tracks regions of interest based on instructor actions to capture dynamic shots. Instructors can adjust the camera behaviors of Stargazer with subtle cues, including gestures and speech, allowing them to fluidly integrate camera control commands into instructional activities. Our user study with six instructors, each teaching a distinct skill, showed that participants could create dynamic tutorial videos with a diverse range of subjects, camera framing, and camera angle combinations using Stargazer. △ Less

Submitted 6 March, 2023; originally announced March 2023.

Comments: To appear in Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI '23), April 23--28, 2023, Hamburg, Germany

GPU-Accelerated DNS of Compressible Turbulent Flows

Authors: Youngdae Kim, Debojyoti Ghosh, Emil M. Constantinescu, Ramesh Balakrishnan

Abstract: This paper explores strategies to transform an existing CPU-based high-performance computational fluid dynamics solver, HyPar, for compressible flow simulations on emerging exascale heterogeneous (CPU+GPU) computing platforms. The scientific motivation for develo** a GPU-enhanced version of HyPar is to simulate canonical turbulent flows at the highest resolution possible on such platforms. We sh… ▽ More This paper explores strategies to transform an existing CPU-based high-performance computational fluid dynamics solver, HyPar, for compressible flow simulations on emerging exascale heterogeneous (CPU+GPU) computing platforms. The scientific motivation for develo** a GPU-enhanced version of HyPar is to simulate canonical turbulent flows at the highest resolution possible on such platforms. We show that optimizing memory operations and thread blocks results in 200x speedup of computationally intensive kernels compared with a CPU core. Using multiple GPUs and CUDA-aware MPI communication, we demonstrate both strong and weak scaling of our GPU-based HyPar implementation on the NVIDIA Volta V100 GPUs. We simulate the decay of homogeneous isotropic turbulence in a triply periodic box on grids with up to $1024^3$ points (5.3 billion degrees of freedom) and on up to 1,024 GPUs. We compare the wall times for CPU-only and CPU+GPU simulations. The results presented in the paper are obtained on the Summit and Lassen supercomputers at Oak Ridge and Lawrence Livermore National Laboratories, respectively. △ Less

Submitted 5 December, 2022; v1 submitted 29 November, 2022; originally announced November 2022.

MSC Class: 76F65; 65Y05; 76F05; 35Q30; 65M06

Differentiable physics-enabled closure modeling for Burgers' turbulence

Authors: Varun Shankar, Vedant Puri, Ramesh Balakrishnan, Romit Maulik, Venkatasubramanian Viswanathan

Abstract: Data-driven turbulence modeling is experiencing a surge in interest following algorithmic and hardware developments in the data sciences. We discuss an approach using the differentiable physics paradigm that combines known physics with machine learning to develop closure models for Burgers' turbulence. We consider the 1D Burgers system as a prototypical test problem for modeling the unresolved ter… ▽ More Data-driven turbulence modeling is experiencing a surge in interest following algorithmic and hardware developments in the data sciences. We discuss an approach using the differentiable physics paradigm that combines known physics with machine learning to develop closure models for Burgers' turbulence. We consider the 1D Burgers system as a prototypical test problem for modeling the unresolved terms in advection-dominated turbulence problems. We train a series of models that incorporate varying degrees of physical assumptions on an a posteriori loss function to test the efficacy of models across a range of system parameters, including viscosity, time, and grid resolution. We find that constraining models with inductive biases in the form of partial differential equations that contain known physics or existing closure approaches produces highly data-efficient, accurate, and generalizable models, outperforming state-of-the-art baselines. Addition of structure in the form of physics information also brings a level of interpretability to the models, potentially offering a step** stone to the future of closure modeling. △ Less

Submitted 23 September, 2022; originally announced September 2022.

Exact Hopfion Vortices in a 3D Heisenberg Ferromagnet

Authors: Radha Balakrishnan, Rossen Dandoloff, Avadh Saxena

Abstract: We find exact static soliton solutions for the unit spin vector field of an inhomogeneous, anisotropic three-dimensional Heisenberg ferromagnet. Each soliton is labeled by two integers $n$ and $m$. It is a (modified) skyrmion in the $z=0$ plane with winding number $n$, which twists out of the plane $m$ times in the $z$-direction to become a 3D soliton. Here $m$ arises due to the periodic boundary… ▽ More We find exact static soliton solutions for the unit spin vector field of an inhomogeneous, anisotropic three-dimensional Heisenberg ferromagnet. Each soliton is labeled by two integers $n$ and $m$. It is a (modified) skyrmion in the $z=0$ plane with winding number $n$, which twists out of the plane $m$ times in the $z$-direction to become a 3D soliton. Here $m$ arises due to the periodic boundary condition at the $z$-boundaries. We use Whitehead's integral expression to find that the Hopf invariant of the soliton is an integer $H =nm$. It represents a hopfion vortex. Plots of the preimages of this topological soliton show that they are either unknots or nontrivial knots, depending on $n$ and $m$. Any pair of preimage curves links $H$ times, corroborating the interpretation of $H$ as a linking number. We also calculate the exact energy of the hopfion vortex, and show that its topological lower bound has a sublinear dependence on $H$. Using Derrick's scaling analysis, we demonstrate that the presence of a spatial inhomogeneity in the anisotropic interaction, which in turn introduces a characteristic length scale in the system, leads to the stability of the hopfion vortex. △ Less

Submitted 2 April, 2023; v1 submitted 15 February, 2022; originally announced February 2022.

Comments: 6 pages, 2 figures

Journal ref: Phys. Lett. A 480, 128975 (2023)

Resistance distance in connected balanced digraphs

Authors: R. Balakrishnan, S. Krishnamoorthy, W. So

Abstract: Let $D = (V, E)$ be a strongly connected and balanced digraph with vertex set $V$ and arc set $E.$ The classical distance $d_{ij}^D$ from $i$ to $j$ in $D$ is the length of a shortest directed path from $i$ to $j$ in $D.$ Let $L$ be the Laplacian matrix of $D$ and $ L^{\dagger} = ( l_{ij}^{\dagger} )$ be the Moore-Penrose inverse of $L.$ The resistance distance from $i$ to $j$ is then defined by… ▽ More Let $D = (V, E)$ be a strongly connected and balanced digraph with vertex set $V$ and arc set $E.$ The classical distance $d_{ij}^D$ from $i$ to $j$ in $D$ is the length of a shortest directed path from $i$ to $j$ in $D.$ Let $L$ be the Laplacian matrix of $D$ and $ L^{\dagger} = ( l_{ij}^{\dagger} )$ be the Moore-Penrose inverse of $L.$ The resistance distance from $i$ to $j$ is then defined by $r_{ij}^D := l_{ii}^{\dagger } + l_{jj}^{\dagger } - 2 l_{ij}^{\dagger }.$ Let $\{ D_1, D_2, ...., D_k \}$ be a sequence of strongly connected balanced digraphs with $D_i \cap D_j$ having at most one vertex in common for all $i \neq j$ and with $r_{ij}^{D_t} \leq d_{ij}^{D_t} \ \forall \ t = 1 \ \mathrm{to} \ k.$ Let $\mathcal{C}$ be a collection of connected, balanced digraphs, each member of which is a finite union of the form $D_1 \cup D_2 \cup ....\cup D_k$ where each $D_i$ is a connected and balanced digraph with $D_{i} \cap ( D_1 \cup D_2 \cup ....\cup D_{i-1} )$ being a single vertex, for all $i,$ $1 < i \leq k.$ In this paper, we show that for any digraph $D$ in $\mathcal{C}$, $r_{ij}^D \leq d_{ij}^D \ (*)$. This is established by partitioning the Laplacian matrix of $D$. This generalizes the main result in [3]. As a corollary, we deduce a simpler proof of the result in [3], namely, that for any directed cactus $D$, the inequality (*) holds. Our results provide an affirmative answer to a well known interesting conjecture ( cf : Conjecture 1.3 ). △ Less

Submitted 31 October, 2022; v1 submitted 27 January, 2022; originally announced January 2022.

MSC Class: 05C50

IMEXLBM 1.0: A Proxy Application based on the Lattice Boltzmann Method for solving Computational Fluid Dynamic problems on GPUs

Authors: Geng Liu, Saumil Patel, Ramesh Balakrishnan, Taehun Lee

Abstract: The US Department of Energy launched the Exascale Computing Project (ECP) in 2016 as part of a coordinated effort to achieve the next generation of high-performance computing (HPC) and to accelerate scientific discovery. The Exascale Proxy Applications Project began within the ECP to: (1) improve the quality of proxies created by the ECP (2) provide small, simplified codes which share important fe… ▽ More The US Department of Energy launched the Exascale Computing Project (ECP) in 2016 as part of a coordinated effort to achieve the next generation of high-performance computing (HPC) and to accelerate scientific discovery. The Exascale Proxy Applications Project began within the ECP to: (1) improve the quality of proxies created by the ECP (2) provide small, simplified codes which share important features of large applications and (3) capture programming methods and styles that drive requirements for compilers and other elements of the tool chain. This article describes one Proxy Application (or "proxy app") suite called IMEXLBM which is an open-source, self-contained code unit, with minimal dependencies, that is capable of running on heterogeneous platforms like those with graphic processing units (GPU) for accelerating the calculation. In particular, we demonstrate functionality by solving a benchmark problem in computational fluid dynamics (CFD) on the ThetaGPU machine at the Argonne Leadership Computing Facility (ALCF). Our method makes use of a domain-decomposition technique in conjunction with the message-passing interface (MPI) standard for distributed memory systems. The OpenMP application programming interface (API) is employed for shared-memory multi-processing and offloading critical kernels to the device (i.e. GPU). We also verify our effort by comparing data generated via CPU-only calculations with data generated with CPU+GPU calculations. While we demonstrate efficacy for single-phase fluid problems, the code-unit is designed to be versatile and enable new physical models that can capture complex phenomena such as two-phase flow with interface capture. △ Less

Submitted 27 January, 2022; originally announced January 2022.

EdgeML: Towards Network-Accelerated Federated Learning over Wireless Edge

Authors: Pinyarash Pinyoanuntapong, Prabhu Janakaraj, Ravikumar Balakrishnan, Minwoo Lee, Chen Chen, Pu Wang

Abstract: Federated learning (FL) is a distributed machine learning technology for next-generation AI systems that allows a number of workers, i.e., edge devices, collaboratively learn a shared global model while kee** their data locally to prevent privacy leakage. Enabling FL over wireless multi-hop networks can democratize AI and make it accessible in a cost-effective manner. However, the noisy bandwidt… ▽ More Federated learning (FL) is a distributed machine learning technology for next-generation AI systems that allows a number of workers, i.e., edge devices, collaboratively learn a shared global model while kee** their data locally to prevent privacy leakage. Enabling FL over wireless multi-hop networks can democratize AI and make it accessible in a cost-effective manner. However, the noisy bandwidth-limited multi-hop wireless connections can lead to delayed and nomadic model updates, which significantly slows down the FL convergence speed. To address such challenges, this paper aims to accelerate FL convergence over wireless edge by optimizing the multi-hop federated networking performance. In particular, the FL convergence optimization problem is formulated as a Markov decision process (MDP). To solve such MDP, multi-agent reinforcement learning (MA-RL) algorithms along with domain-specific action space refining schemes are developed, which online learn the delay-minimum forwarding paths to minimize the model exchange latency between the edge devices (i.e., workers) and the remote server. To validate the proposed solutions, FedEdge is developed and implemented, which is the first experimental framework in the literature for FL over multi-hop wireless edge computing networks. FedEdge allows us to fast prototype, deploy, and evaluate novel FL algorithms along with RL-based system optimization methods in real wireless devices. Moreover, a physical experimental testbed is implemented by customizing the widely adopted Linux wireless routers and ML computing nodes.Finally, our experimentation results on the testbed show that the proposed network-accelerated FL system can practically and significantly improve FL convergence speed, compared to the FL system empowered by the production-grade commercially available wireless networking protocol, BATMAN-Adv. △ Less

Submitted 31 May, 2022; v1 submitted 14 October, 2021; originally announced November 2021.

Comments: 14 pages, submitted to Journal of Computer Networks (ComNet) 2022

Protecting Anonymous Speech: A Generative Adversarial Network Methodology for Removing Stylistic Indicators in Text

Authors: Rishi Balakrishnan, Stephen Sloan, Anil Aswani

Abstract: With Internet users constantly leaving a trail of text, whether through blogs, emails, or social media posts, the ability to write and protest anonymously is being eroded because artificial intelligence, when given a sample of previous work, can match text with its author out of hundreds of possible candidates. Existing approaches to authorship anonymization, also known as authorship obfuscation,… ▽ More With Internet users constantly leaving a trail of text, whether through blogs, emails, or social media posts, the ability to write and protest anonymously is being eroded because artificial intelligence, when given a sample of previous work, can match text with its author out of hundreds of possible candidates. Existing approaches to authorship anonymization, also known as authorship obfuscation, often focus on protecting binary demographic attributes rather than identity as a whole. Even those that do focus on obfuscating identity require manual feedback, lose the coherence of the original sentence, or only perform well given a limited subset of authors. In this paper, we develop a new approach to authorship anonymization by constructing a generative adversarial network that protects identity and optimizes for three different losses corresponding to anonymity, fluency, and content preservation. Our fully automatic method achieves comparable results to other methods in terms of content preservation and fluency, but greatly outperforms baselines in regards to anonymization. Moreover, our approach is able to generalize well to an open-set context and anonymize sentences from authors it has not encountered before. △ Less

Submitted 18 October, 2021; originally announced October 2021.

Sim-to-Real Transfer in Multi-agent Reinforcement Networking for Federated Edge Computing

Authors: Pinyarash Pinyoanuntapong, Tagore Pothuneedi, Ravikumar Balakrishnan, Minwoo Lee, Chen Chen, Pu Wang

Abstract: Federated Learning (FL) over wireless multi-hop edge computing networks, i.e., multi-hop FL, is a cost-effective distributed on-device deep learning paradigm. This paper presents FedEdge simulator, a high-fidelity Linux-based simulator, which enables fast prototy**, sim-to-real code, and knowledge transfer for multi-hop FL systems. FedEdge simulator is built on top of the hardware-oriented FedEd… ▽ More Federated Learning (FL) over wireless multi-hop edge computing networks, i.e., multi-hop FL, is a cost-effective distributed on-device deep learning paradigm. This paper presents FedEdge simulator, a high-fidelity Linux-based simulator, which enables fast prototy**, sim-to-real code, and knowledge transfer for multi-hop FL systems. FedEdge simulator is built on top of the hardware-oriented FedEdge experimental framework with a new extension of the realistic physical layer emulator. This emulator exploits trace-based channel modeling and dynamic link scheduling to minimize the reality gap between the simulator and the physical testbed. Our initial experiments demonstrate the high fidelity of the FedEdge simulator and its superior performance on sim-to-real knowledge transfer in reinforcement learning-optimized multi-hop FL. △ Less

Submitted 17 October, 2021; originally announced October 2021.

Comments: 5 pages

AutoTune: Improving End-to-end Performance and Resource Efficiency for Microservice Applications

Authors: Michael Alan Chang, Aurojit Panda, Hantao Wang, Yuancheng Tsai, Rahul Balakrishnan, Scott Shenker

Abstract: Most large web-scale applications are now built by composing collections (from a few up to 100s or 1000s) of microservices. Operators need to decide how many resources are allocated to each microservice, and these allocations can have a large impact on application performance. Manually determining allocations that are both cost-efficient and meet performance requirements is challenging, even for e… ▽ More Most large web-scale applications are now built by composing collections (from a few up to 100s or 1000s) of microservices. Operators need to decide how many resources are allocated to each microservice, and these allocations can have a large impact on application performance. Manually determining allocations that are both cost-efficient and meet performance requirements is challenging, even for experienced operators. In this paper we present AutoTune, an end-to-end tool that automatically minimizes resource utilization while maintaining good application performance. △ Less

Submitted 28 June, 2021; v1 submitted 18 June, 2021; originally announced June 2021.

MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

Authors: Taojiannan Yang, Sijie Zhu, Matias Mendieta, Pu Wang, Ravikumar Balakrishnan, Minwoo Lee, Tao Han, Mubarak Shah, Chen Chen

Abstract: Most existing deep neural networks are static, which means they can only do inference at a fixed complexity. But the resource budget can vary substantially across different devices. Even on a single device, the affordable budget can change with different scenarios, and repeatedly training networks for each required budget would be incredibly expensive. Therefore, in this work, we propose a general… ▽ More Most existing deep neural networks are static, which means they can only do inference at a fixed complexity. But the resource budget can vary substantially across different devices. Even on a single device, the affordable budget can change with different scenarios, and repeatedly training networks for each required budget would be incredibly expensive. Therefore, in this work, we propose a general method called MutualNet to train a single network that can run at a diverse set of resource constraints. Our method trains a cohort of model configurations with various network widths and input resolutions. This mutual learning scheme not only allows the model to run at different width-resolution configurations but also transfers the unique knowledge among these configurations, hel** the model to learn stronger representations overall. MutualNet is a general training methodology that can be applied to various network structures (e.g., 2D networks: MobileNets, ResNet, 3D networks: SlowFast, X3D) and various tasks (e.g., image classification, object detection, segmentation, and action recognition), and is demonstrated to achieve consistent improvements on a variety of datasets. Since we only train the model once, it also greatly reduces the training cost compared to independently training several models. Surprisingly, MutualNet can also be used to significantly boost the performance of a single network, if dynamic resource constraint is not a concern. In summary, MutualNet is a unified method for both static and adaptive, 2D and 3D networks. Codes and pre-trained models are available at \url{https://github.com/taoyang1122/MutualNet}. △ Less

Submitted 30 December, 2021; v1 submitted 14 May, 2021; originally announced May 2021.

Comments: Extended version of arXiv:1909.12978. Updated analyses and results. Accepted to TPAMI

FuSSI-Net: Fusion of Spatio-temporal Skeletons for Intention Prediction Network

Authors: Francesco Piccoli, Rajarathnam Balakrishnan, Maria Jesus Perez, Moraldeepsingh Sachdeo, Carlos Nunez, Matthew Tang, Kajsa Andreasson, Kalle Bjurek, Ria Dass Raj, Ebba Davidsson, Colin Eriksson, Victor Hagman, Jonas Sjoberg, Ying Li, L. Srikar Muppirisetty, Sohini Roychowdhury

Abstract: Pedestrian intention recognition is very important to develop robust and safe autonomous driving (AD) and advanced driver assistance systems (ADAS) functionalities for urban driving. In this work, we develop an end-to-end pedestrian intention framework that performs well on day- and night- time scenarios. Our framework relies on objection detection bounding boxes combined with skeletal features of… ▽ More Pedestrian intention recognition is very important to develop robust and safe autonomous driving (AD) and advanced driver assistance systems (ADAS) functionalities for urban driving. In this work, we develop an end-to-end pedestrian intention framework that performs well on day- and night- time scenarios. Our framework relies on objection detection bounding boxes combined with skeletal features of human pose. We study early, late, and combined (early and late) fusion mechanisms to exploit the skeletal features and reduce false positives as well to improve the intention prediction performance. The early fusion mechanism results in AP of 0.89 and precision/recall of 0.79/0.89 for pedestrian intention classification. Furthermore, we propose three new metrics to properly evaluate the pedestrian intention systems. Under these new evaluation metrics for the intention prediction, the proposed end-to-end network offers accurate pedestrian intention up to half a second ahead of the actual risky maneuver. △ Less

Submitted 15 May, 2020; originally announced May 2020.

Comments: 5 pages, 6 figures, 5 tables, IEEE Asilomar SSC

On the Vasconcelos inequality for the fiber multiplicity of modules

Authors: R. Balakrishnan, A. V. Jayanthan

Abstract: Let $(R,\mathfrak{m})$ be a Noetherian local ring of dimension $d>0$ with infinite residue field. Let $M$ be a finitely generated proper $R$-submodule of a free $R$-module $F$ with $\ell (F/M) < \infty$ and having rank $r$. In this article, we study the fiber multiplicity $f_0(M)$ of the module $M$. We prove that if $(R,\mathfrak{m})$ is a two dimensional Cohen-Macaulay local ring, then… ▽ More Let $(R,\mathfrak{m})$ be a Noetherian local ring of dimension $d>0$ with infinite residue field. Let $M$ be a finitely generated proper $R$-submodule of a free $R$-module $F$ with $\ell (F/M) < \infty$ and having rank $r$. In this article, we study the fiber multiplicity $f_0(M)$ of the module $M$. We prove that if $(R,\mathfrak{m})$ is a two dimensional Cohen-Macaulay local ring, then $f_0(M)\le br_1(M)-br_0(M)+\ell (F/M)+μ(M)-r$, where $br_i(M)$ denotes the $i^{th}$ Buchsbaum-Rim coefficient of $M$. △ Less

Submitted 4 April, 2018; originally announced April 2018.

Comments: 18 pages. (To appear in Comm. Algebra)

MSC Class: 13D40; 13A30

Augmenting Mobile Phone Interaction with Face-Engaged Gestures

Authors: Jian Zhao, Ricardo Jota, Daniel Wigdor, Ravin Balakrishnan

Abstract: The movement of a user's face, easily detected by a smartphone's front camera, is an underexploited input modality for mobile interactions. We introduce three sets of face-engaged interaction techniques for augmenting the traditional mobile inputs, which leverages the combination of the head movements with touch gestures and device motions, all sensed via the phone's built-in sensors. We systemati… ▽ More The movement of a user's face, easily detected by a smartphone's front camera, is an underexploited input modality for mobile interactions. We introduce three sets of face-engaged interaction techniques for augmenting the traditional mobile inputs, which leverages the combination of the head movements with touch gestures and device motions, all sensed via the phone's built-in sensors. We systematically present the space of design considerations for mobile interactions using one or more of the three input modalities (i.e., touch, motion, and head). The additional affordances of the proposed techniques expand the mobile interaction vocabulary, and can facilitate unique usage scenarios such as one-hand or touch-free interaction. An initial evaluation was conducted and users had positive reactions to the new techniques, indicating the promise of an intuitive and convenient user experience. △ Less

Submitted 1 October, 2016; originally announced October 2016.

Comments: 9 pages

ACM Class: H.5.2

A Dataset of Naturally Occurring, Whole-Body Background Activity to Reduce Gesture Conflicts

Authors: Dustin Freeman, Ricardo Jota, Daniel Vogel, Daniel Wigdor, Ravin Balakrishnan

Abstract: In real settings, natural body movements can be erroneously recognized by whole-body input systems as explicit input actions. We call body activity not intended as input actions "background activity." We argue that understanding background activity is crucial to the success of always-available whole-body input in the real world. To operationalize this argument, we contribute a reusable study metho… ▽ More In real settings, natural body movements can be erroneously recognized by whole-body input systems as explicit input actions. We call body activity not intended as input actions "background activity." We argue that understanding background activity is crucial to the success of always-available whole-body input in the real world. To operationalize this argument, we contribute a reusable study methodology and software tools to generate standardized background activity datasets composed of data from multiple Kinect cameras, a Vicon tracker, and two high-definition video cameras. Using our methodology, we create an example background activity dataset for a television-oriented living room setting. We use this dataset to demonstrate how it can be used to redesign a gestural interaction vocabulary to minimize conflicts with the real world. The software tools and initial living room dataset are publicly available (http://www.dgp.toronto.edu/~dustin/backgroundactivity/). △ Less

Submitted 21 September, 2015; originally announced September 2015.

ACM Class: H.5.m

Solitons in a hard-core bosonic system: Gross-Pitaevskii type and beyond

Authors: Radha Balakrishnan, Indubala I Satija

Abstract: A unified formulation that obtains solitary waves for various background densities in the Bose-Einstein condensate of a system of hard-core bosons with nearest neighbor attractive interactions is presented. In general, two species of solitons appear: A nonpersistent (NP) type that fully delocalizes at its maximum speed, and a persistent (P) type that survives even at its maximum speed, and trans… ▽ More A unified formulation that obtains solitary waves for various background densities in the Bose-Einstein condensate of a system of hard-core bosons with nearest neighbor attractive interactions is presented. In general, two species of solitons appear: A nonpersistent (NP) type that fully delocalizes at its maximum speed, and a persistent (P) type that survives even at its maximum speed, and transforms into a periodic train of solitons above this speed. When the background condensate density is nonzero, both species coexist, the soliton is associated with a constant intrinsic frequency, and its maximum speed is the speed of sound. In contrast, when the background condensate density is zero, the system has neither a fixed frequency, nor a speed of sound. Here, the maximum soliton speed depends on the frequency, which can be tuned to lead to a cross-over between the NP-type and the P-type at a certain critical frequency, determined by the energy parameters of the system. We provide a single functional form for the soliton profile, from which diverse characteristics for various background densities can be obtained. Using the map** to spin systems enables us to characterize the corresponding class of magnetic solitons in Heisenberg spin chains with different types of anisotropy, in a unified fashion. △ Less

Submitted 20 April, 2014; originally announced April 2014.

RAProp: Ranking Tweets by Exploiting the Tweet/User/Web Ecosystem and Inter-Tweet Agreement

Authors: Srijith Ravikumar, Kartik Talamadupula, Raju Balakrishnan, Subbarao Kambhampati

Abstract: The increasing popularity of Twitter renders improved trustworthiness and relevance assessment of tweets much more important for search. However, given the limitations on the size of tweets, it is hard to extract measures for ranking from the tweets' content alone. We present a novel ranking method, called RAProp, which combines two orthogonal measures of relevance and trustworthiness of a tweet.… ▽ More The increasing popularity of Twitter renders improved trustworthiness and relevance assessment of tweets much more important for search. However, given the limitations on the size of tweets, it is hard to extract measures for ranking from the tweets' content alone. We present a novel ranking method, called RAProp, which combines two orthogonal measures of relevance and trustworthiness of a tweet. The first, called Feature Score, measures the trustworthiness of the source of the tweet. This is done by extracting features from a 3-layer twitter ecosystem, consisting of users, tweets and the pages referred to in the tweets. The second measure, called agreement analysis, estimates the trustworthiness of the content of the tweet, by analyzing how and whether the content is independently corroborated by other tweets. We view the candidate result set of tweets as the vertices of a graph, with the edges measuring the estimated agreement between each pair of tweets. The feature score is propagated over this agreement graph to compute the top-k tweets that have both trustworthy sources and independent corroboration. The evaluation of our method on 16 million tweets from the TREC 2011 Microblog Dataset shows that for top-30 precision we achieve 53% higher than current best performing method on the Dataset and over 300% over current Twitter Search. We also present a detailed internal empirical evaluation of RAProp in comparison to several alternative approaches proposed by us. △ Less

Submitted 10 August, 2013; originally announced August 2013.

Comments: 11 pages

Real-Time Bid Optimization for Group-Buying Ads

Authors: Raju Balakrishnan, Rushi P Bhatt

Abstract: Group-buying ads seeking a minimum number of customers before the deal expiry are increasingly used by the daily-deal providers. Unlike the traditional web ads, the advertiser's profits for group-buying ads depends on the time to expiry and additional customers needed to satisfy the minimum group size. Since both these quantities are time-dependent, optimal bid amounts to maximize profits change w… ▽ More Group-buying ads seeking a minimum number of customers before the deal expiry are increasingly used by the daily-deal providers. Unlike the traditional web ads, the advertiser's profits for group-buying ads depends on the time to expiry and additional customers needed to satisfy the minimum group size. Since both these quantities are time-dependent, optimal bid amounts to maximize profits change with every impression. Consequently, traditional static bidding strategies are far from optimal. Instead, bid values need to be optimized in real-time to maximize expected bidder profits. This online optimization of deal profits is made possible by the advent of ad exchanges offering real-time (spot) bidding. To this end, we propose a real-time bidding strategy for group-buying deals based on the online optimization of bid values. We derive the expected bidder profit of deals as a function of the bid amounts, and dynamically vary bids to maximize profits. Further, to satisfy time constraints of the online bidding, we present methods of minimizing computation timings. Subsequently, we derive the real time ad selection, admissibility, and real time bidding of the traditional ads as the special cases of the proposed method. We evaluate the proposed bidding, selection and admission strategies on a multi-million click stream of 935 ads. The proposed real-time bidding, selection and admissibility show significant profit increases over the existing strategies. Further the experiments illustrate the robustness of the bidding and acceptable computation timings. △ Less

Submitted 3 June, 2012; originally announced June 2012.

Ranking Tweets Considering Trust and Relevance

Authors: Srijith Ravikumar, Raju Balakrishnan, Subbarao Kambhampati

Abstract: The increasing popularity of Twitter and other microblogs makes improved trustworthiness and relevance assessment of microblogs evermore important. We propose a method of ranking of tweets considering trustworthiness and content based popularity. The analysis of trustworthiness and popularity exploits the implicit relationships between the tweets. We model microblog ecosystem as a three-layer grap… ▽ More The increasing popularity of Twitter and other microblogs makes improved trustworthiness and relevance assessment of microblogs evermore important. We propose a method of ranking of tweets considering trustworthiness and content based popularity. The analysis of trustworthiness and popularity exploits the implicit relationships between the tweets. We model microblog ecosystem as a three-layer graph consisting of : (i) users (ii) tweets and (iii) web pages. We propose to derive trust and popularity scores of entities in these three layers, and propagate the scores to tweets considering the inter-layer relations. Our preliminary evaluations show improvement in precision and trustworthiness over the baseline methods and acceptable computation timings. △ Less

Submitted 31 March, 2012; originally announced April 2012.

Comments: four pages short paper

Quantum Dynamics of Solitons in Strongly Interacting Systems on Optical Lattices

Authors: Chester P. Rubbo, Indubala I. Satija, William P. Reinhardt, Radha Balakrishnan, Ana Maria Rey, Salvatore R. Manmana

Abstract: Mean-field dynamics of strongly interacting bosons described by hard core bosons with nearest-neighbor attraction has been shown to support two species of solitons: one of Gross-Pitaevskii (GP-type) where the condensate fraction remains dark and a novel non-Gross-Pitaevskii-type (non-GP-type) characterized by brightening of the condensate fraction. Here we study the effects of quantum fluctuations… ▽ More Mean-field dynamics of strongly interacting bosons described by hard core bosons with nearest-neighbor attraction has been shown to support two species of solitons: one of Gross-Pitaevskii (GP-type) where the condensate fraction remains dark and a novel non-Gross-Pitaevskii-type (non-GP-type) characterized by brightening of the condensate fraction. Here we study the effects of quantum fluctuations on these solitons using the adaptive time-dependent density matrix renormalization group method, which takes into account the effect of strong correlations. We use local observables as the density, condensate density and correlation functions as well as the entanglement entropy to characterize the stability of the initial states. We find both species of solitons to be stable under quantum evolution for a finite duration, their tolerance to quantum fluctuations being enhanced as the width of the soliton increases. We describe possible experimental realizations in atomic Bose Einstein Condensates, polarized degenerate Fermi gases, and in systems of polar molecules on optical lattices. △ Less

Submitted 22 May, 2012; v1 submitted 15 February, 2012; originally announced February 2012.

Journal ref: Phys. Rev. A 85, 053617 (2012)

Click Efficiency: A Unified Optimal Ranking for Online Ads and Documents

Authors: Raju Balakrishnan, Subbarao Kambhampati

Abstract: Traditionally the probabilistic ranking principle is used to rank the search results while the ranking based on expected profits is used for paid placement of ads. These rankings try to maximize the expected utilities based on the user click models. Recent empirical analysis on search engine logs suggests a unified click models for both ranked ads and search results. The segregated view of documen… ▽ More Traditionally the probabilistic ranking principle is used to rank the search results while the ranking based on expected profits is used for paid placement of ads. These rankings try to maximize the expected utilities based on the user click models. Recent empirical analysis on search engine logs suggests a unified click models for both ranked ads and search results. The segregated view of document and ad rankings does not consider this commonality. Further, the used models consider parameters of (i) probability of the user abandoning browsing results (ii) perceived relevance of result snippets. But how to consider them for improved ranking is unknown currently. In this paper, we propose a generalized ranking function---namely "Click Efficiency (CE)"---for documents and ads based on empirically proven user click models. The ranking considers parameters (i) and (ii) above, optimal and has the same time complexity as sorting. To exploit its generality, we examine the reduced forms of CE ranking under different assumptions enumerating a hierarchy of ranking functions. Some of the rankings in the hierarchy are currently used ad and document ranking functions; while others suggest new rankings. While optimality of ranking is sufficient for document ranking, applying CE ranking to ad auctions requires an appropriate pricing mechanism. We incorporate a second price based pricing mechanism with the proposed ranking. Our analysis proves several desirable properties including revenue dominance over VCG for the same bid vector and existence of a Nash Equilibrium in pure strategies. The equilibrium is socially optimal, and revenue equivalent to the truthful VCG equilibrium. Further, we relax the independence assumption in CE ranking and analyze the diversity ranking problem. We show that optimal diversity ranking is NP-Hard in general, and that a constant time approximation is unlikely. △ Less

Submitted 8 July, 2011; originally announced July 2011.

Comments: Twenty Six Pages, Two Figures, Six Theorems. Initial Version Published in Workshop on Web and Databases 2008

ACM Class: H.3.3

A sharp lower bound for the Wiener index of a graph

Authors: R. Balakrishnan, N. Sridharan, K. V. Iyer

Abstract: Given a simple connected undirected graph G, the Wiener index W(G) of G is defined as half the sum of the distances over all pairs of vertices of G. In practice, G corresponds to what is known as the molecular graph of an organic compound. We obtain a sharp lower bound for W(G) of an arbitrary graph in terms of the order, size and diameter of G. Given a simple connected undirected graph G, the Wiener index W(G) of G is defined as half the sum of the distances over all pairs of vertices of G. In practice, G corresponds to what is known as the molecular graph of an organic compound. We obtain a sharp lower bound for W(G) of an arbitrary graph in terms of the order, size and diameter of G. △ Less

Submitted 24 August, 2010; originally announced August 2010.

Comments: Accepted for publication in Ars Combinatoria

Journal ref: Ars Combinatoria,v.XCVII,Oct. 2010

Other incarnations of the Gross-Pitaevskii dark soliton

Authors: Indubala I Satija, Radha Balakrishnan

Abstract: We show that the dark soliton of the Gross-Pitaevskii equation (GPE) that describes the Bose-Einstein condensate (BEC) density of a system of weakly repulsive bosons, also describes that of a system of strongly repulsive hard core bosons at half filling. As a consequence of this, the GPE soliton gets related to the magnetic soliton in an easy-plane ferromagnet, where it describes the square of the… ▽ More We show that the dark soliton of the Gross-Pitaevskii equation (GPE) that describes the Bose-Einstein condensate (BEC) density of a system of weakly repulsive bosons, also describes that of a system of strongly repulsive hard core bosons at half filling. As a consequence of this, the GPE soliton gets related to the magnetic soliton in an easy-plane ferromagnet, where it describes the square of the in-plane magnetization of the system. These relationships are shown to be useful in understanding various characteristics of solitons in these distinct many-body systems. △ Less

Submitted 21 June, 2010; originally announced June 2010.

The Gross-Pitaevskii Soliton: Relating Weakly and Strongly Repulsive Bosonic condensates and the magnetic soliton

Authors: Indubala I Satija, Radha Balakrishnan

Abstract: We show that the dark soliton of the Gross-Pitaevskii equation (GPE) that describes the Bose-Einstein condensate (BEC) density of a system of weakly repulsive bosons, also describes that of a system of strongly repulsive hard core bosons at half filling. This connection establishes a relationship between the GPE soliton and the magnetic soliton of an easy-plane ferromagnet, where the BEC dens… ▽ More We show that the dark soliton of the Gross-Pitaevskii equation (GPE) that describes the Bose-Einstein condensate (BEC) density of a system of weakly repulsive bosons, also describes that of a system of strongly repulsive hard core bosons at half filling. This connection establishes a relationship between the GPE soliton and the magnetic soliton of an easy-plane ferromagnet, where the BEC density relates to the square of the in-plane magnetization of the system. This map** between well known solitons in two distinct physical systems provides an intuitive understanding of various characteristics of the solitons. △ Less

Submitted 13 April, 2010; originally announced April 2010.

Particle-Hole Asymmetry and Brightening of Solitons in A Strongly Repulsive BEC

Authors: Radha Balakrishnan, Indubala I. Satija, Charles W. Clark

Abstract: We study solitary wave propagation in the condensate of a system of hard-core bosons with nearest-neighbor interactions. For this strongly repulsive system, the evolution equation for the condensate order parameter of the system, obtained using spin coherent state averages is different from the usual Gross-Pitaevskii equation (GPE). The system is found to support two kinds of solitons when there… ▽ More We study solitary wave propagation in the condensate of a system of hard-core bosons with nearest-neighbor interactions. For this strongly repulsive system, the evolution equation for the condensate order parameter of the system, obtained using spin coherent state averages is different from the usual Gross-Pitaevskii equation (GPE). The system is found to support two kinds of solitons when there is a particle-hole imbalance: a dark soliton that dies out as the velocity approaches the sound velocity, and a new type of soliton which brightens and persists all the way up to the sound velocity, transforming into a periodic wave train at supersonic speed. Analogous to the GPE soliton, the energy-momentum dispersion for both solitons is characterized by Lieb II modes. △ Less

Submitted 15 November, 2009; v1 submitted 26 August, 2009; originally announced August 2009.

Comments: Accepted for publication in PRL, Nov 12, 2009

Journal ref: Phys.Rev.Lett.103:230403,2009

Symmetry-Breaking and Symmetry-Restoring Dynamics of a Mixture of Bose-Einstein Condensates in a Double Well

Authors: Indubala I Satija, Philip Naudus, Radha Balakrishnan, Jeffrey Heward, Mark Edwards, and Charles W Clark

Abstract: We study the coherent nonlinear tunneling dynamics of a binary mixture of Bose-Einstein condensates in a double-well potential. We demonstrate the existence of a new type of mode associated with the "swap**" of the two species in the two wells of the potential. In contrast to the symmetry breaking macroscopic quantum self-trap** (MQST) solutions, the swap** modes correspond to the tunnelin… ▽ More We study the coherent nonlinear tunneling dynamics of a binary mixture of Bose-Einstein condensates in a double-well potential. We demonstrate the existence of a new type of mode associated with the "swap**" of the two species in the two wells of the potential. In contrast to the symmetry breaking macroscopic quantum self-trap** (MQST) solutions, the swap** modes correspond to the tunneling dynamics that preserves the symmetry of the double well potential. As a consequence of two distinct types of broken symmetry MQST phases where the two species localize in the different potential welils or coexist in the same well, the corresponding symmetry restoring swap** modes result in dynamics where the the two species either avoid or chase each other. In view of the possibility to control the interaction between the species, the binary mixture offers a very robust system to observe these novel effects as well as the phenomena of Josephson oscillations and pi-modes △ Less

Submitted 12 November, 2008; originally announced November 2008.

Effect of conformations on charge transport in a thin elastic tube

Authors: Radha Balakrishnan, Rossen Dandoloff

Abstract: We study the effect of conformations on charge transport in a thin elastic tube. Using the Kirchhoff model for a tube with any given Poisson ratio, cross-sectional shape and intrinsic twist, we obtain a class of exact solutions for its conformation. The tube's torsion is found in terms of its intrinsic twist and its Poisson ratio, while its curvature satisfies a nonlinear differential equation w… ▽ More We study the effect of conformations on charge transport in a thin elastic tube. Using the Kirchhoff model for a tube with any given Poisson ratio, cross-sectional shape and intrinsic twist, we obtain a class of exact solutions for its conformation. The tube's torsion is found in terms of its intrinsic twist and its Poisson ratio, while its curvature satisfies a nonlinear differential equation which supports exact {\it periodic} solutions in the form of Jacobi elliptic functions, which we call {\it conformon lattice} solutions. These solutions typically describe conformations with loops. Each solution induces a corresponding quantum effective {\it periodic} potential in the Schrödinger equation for an electron in the tube. The wave function describes the delocalization of the electron along the central axis of the tube. We discuss some possible applications of this novel mechanism of charge transport. △ Less

Submitted 24 March, 2008; originally announced March 2008.

Journal ref: Nonlinearity, 21 (2008) p.1

Optical phase shifts and diabolic topology in Mobius-type strips

Authors: Indubala I Satija, Radha Balakrishnan

Abstract: We compute the optical phase shifts between the left and the right-circularly polarized light after it traverses non-planar cyclic paths described by the boundary curves of closed twisted strips. The evolution of the electric field along the curved path of a light ray is described by the Fermi-Walker transport law which is mapped to a Schrödinger equation. The effective quantum Hamiltonian of th… ▽ More We compute the optical phase shifts between the left and the right-circularly polarized light after it traverses non-planar cyclic paths described by the boundary curves of closed twisted strips. The evolution of the electric field along the curved path of a light ray is described by the Fermi-Walker transport law which is mapped to a Schrödinger equation. The effective quantum Hamiltonian of the system has eigenvalues equal to $0, \pm κ$, where $κ$ is the local curvature of the path. The inflexion points of the twisted strips correspond to the vanishing of the curvature and manifest themselves as the diabolic crossings of the quantum Hamiltonian. For the Möbius loops, the critical width where the diabolic geometry resides also corresponds to the characteristic width where the optical phase shift is minimal. In our detailed study of various twisted strips, this intriguing property singles out the M"{o}bius geometry. △ Less

Submitted 16 January, 2007; originally announced January 2007.

Gauge Invariant Geometry of Closed Space Curves: Applications to Boundary Curves of Mobius-type Strips

Authors: Radha Balakrishnan, Indubala I Satija

Abstract: We derive gauge-invariant expressions for the twist $Tw$ and the linking number $Lk$ of a closed space curve, that are independent of the frame used to describe the curve, and hence characterize the intrinsic geometry of the curve. We are thus led to a {\it frame-independent} version of the Călugăreanu-White-Fuller theorem $Lk =Tw + Wr$ for a curve, where $Wr$ is the writhe of the curve. The gau… ▽ More We derive gauge-invariant expressions for the twist $Tw$ and the linking number $Lk$ of a closed space curve, that are independent of the frame used to describe the curve, and hence characterize the intrinsic geometry of the curve. We are thus led to a {\it frame-independent} version of the Călugăreanu-White-Fuller theorem $Lk =Tw + Wr$ for a curve, where $Wr$ is the writhe of the curve. The gauge-invariant twist and writhe are related to two types of geometric phases associated with the curve. As an application, we study the geometry of the boundary curves of closed twisted strips. Interestingly, the Möbius strip geometry is singled out by a characteristic maximum that appears in the geometric phases, at a certain critical width of the strip. △ Less

Submitted 15 July, 2005; originally announced July 2005.

Quantum effective potential, electron transport and conformons in biopolymers

Authors: Rossen Dandoloff, Radha Balakrishnan

Abstract: In the Kirchhoff model of a biopolymer, conformation dynamics can be described in terms of solitary waves, for certain special cross-section asymmetries. Applying this to the problem of electron transport, we show that the quantum effective potential arising due to the bends and twists of the polymer enables us to formalize and quantify the concept of a {\it conformon} that has been hypothesized… ▽ More In the Kirchhoff model of a biopolymer, conformation dynamics can be described in terms of solitary waves, for certain special cross-section asymmetries. Applying this to the problem of electron transport, we show that the quantum effective potential arising due to the bends and twists of the polymer enables us to formalize and quantify the concept of a {\it conformon} that has been hypothesized in biology. Its connection to the soliton solution of the cubic nonlinear Schrödinger equation emerges in a natural fashion. △ Less

Submitted 29 May, 2005; originally announced May 2005.

Comments: to appear in J. Phys. A

Origin of Low-Frequency Negative Transconductance Dispersion in p-HEMT

Authors: V. R. Balakrishnan, Vikram Kumar, Subhasis Ghosh

Abstract: Measurements of low-frequency transconductance dispersion at different temperatures and conductance deep level transient spectroscopic(CDLTS) studies of an AlGaAs/InGaAs pseudomorphic HEMT were carried out. The experimental results show the presence of defect states at the AlGaAs/InGaAs hetero-interface. A mobility degradation model was developed to explain the low frequency negative transconduc… ▽ More Measurements of low-frequency transconductance dispersion at different temperatures and conductance deep level transient spectroscopic(CDLTS) studies of an AlGaAs/InGaAs pseudomorphic HEMT were carried out. The experimental results show the presence of defect states at the AlGaAs/InGaAs hetero-interface. A mobility degradation model was developed to explain the low frequency negative transconductance dispersion as well as the apparent hole-like peaks observed in the CDLTS spectra. This model incorporates a time dependent change in 2DEG mobility due to ionised impurity scattering by the remaining charge states at the adjoining AlGaAs/InGaAs hetero-interface. △ Less

Submitted 27 March, 2005; originally announced March 2005.

Geometric Phase and Classical-Quantum Correspondence

Authors: Indubala I. Satija, Radha Balakrishnan

Abstract: We study the geometric phase factors underlying the classical and the corresponding quantum dynamics of a driven nonlinear oscillator exhibiting chaotic dynamics. For the classical problem, we compute the geometric phase factors associated with the phase space trajectories using Frenet-Serret formulation. For the corresponding quantum problem, the geometric phase associated with the time evoluti… ▽ More We study the geometric phase factors underlying the classical and the corresponding quantum dynamics of a driven nonlinear oscillator exhibiting chaotic dynamics. For the classical problem, we compute the geometric phase factors associated with the phase space trajectories using Frenet-Serret formulation. For the corresponding quantum problem, the geometric phase associated with the time evolution of the wave function is computed. Our studies suggest that the classical geometric phase may be related to the the difference in the quantum geometric phases between two neighboring eigenstates. △ Less

Submitted 5 March, 2004; originally announced March 2004.

Comments: Copy with higher resolution figures can be obtained from http://physics.gmu.edu/~isatija by clicking on publications. to appear in the Yukawa Institute conference proceedings, {\it Quantum Mechanics and Chaos: From Fundamental Problems through Nano-Science} (2003)

Geometric Phase in a Bose-Einstein Josephson Junction

Authors: Radha Balakrishnan, Mitaxi Mehta

Abstract: We calculate the geometric phase associated with the time evolution of the wave function of a Bose-Einstein condensate system in a double-well trap by using a model for tunneling between the wells. For a cyclic evolution, this phase is shown to be half the solid angle subtended by the evolution of a unit vector whose z component and azimuthal angle are given by the population difference and phas… ▽ More We calculate the geometric phase associated with the time evolution of the wave function of a Bose-Einstein condensate system in a double-well trap by using a model for tunneling between the wells. For a cyclic evolution, this phase is shown to be half the solid angle subtended by the evolution of a unit vector whose z component and azimuthal angle are given by the population difference and phase difference between the two condensates. For a non-cyclic evolution an additional phase term arises. We show that the geometric phase can also be obtained by map** the tunneling equations onto the equations os a space curve. The importance of a geometric phase in the context of some recent experiments is pointed out. △ Less

Submitted 4 July, 2003; v1 submitted 3 July, 2003; originally announced July 2003.

Comments: 32 pages 9 figures

Nonlinear elastodynamics and energy transport in biopolymers

Authors: Radha Balakrishnan, Rossen Dandoloff

Abstract: We present an exact analysis of the intrinsic dynamics of two types of biopolymers, described by the wormlike chain (WLC) and the wormlike rod chain (WLRC) models, respectively. We map each model to a classical field theory for an appropriate unit vector field defined on the space curve formed by the axis of the biopolymer. Conformation dynamics and energy transport are shown to be described, re… ▽ More We present an exact analysis of the intrinsic dynamics of two types of biopolymers, described by the wormlike chain (WLC) and the wormlike rod chain (WLRC) models, respectively. We map each model to a classical field theory for an appropriate unit vector field defined on the space curve formed by the axis of the biopolymer. Conformation dynamics and energy transport are shown to be described, respectively, by the soliton-bearing Landau-Lifshitz equation and the nonlinear Schrödinger equation. This enables us to identify the soliton representing the ``conformon'' hypothesized in biopolymers. △ Less

Submitted 30 April, 2003; originally announced April 2003.

Comments: 4 pages

Anholonomy and Geometrical Localization in Dynamical Systems

Authors: Radha Balakrishnan, Indubala Satija

Abstract: We characterize the geometrical and topological aspects of a dynamical system by associating a geometric phase with a phase space trajectory. Using the example of a nonlinear driven damped oscillator, we show that this phase is resilient to fluctuations, responds to all bifurcations in the system, and also finds new geometric transitions. Enriching the phase space description is a novel phenomen… ▽ More We characterize the geometrical and topological aspects of a dynamical system by associating a geometric phase with a phase space trajectory. Using the example of a nonlinear driven damped oscillator, we show that this phase is resilient to fluctuations, responds to all bifurcations in the system, and also finds new geometric transitions. Enriching the phase space description is a novel phenomenon of ``geometrical localization'' which manifests itself as a significant deviation from planar dynamics over a short time interval. △ Less

Submitted 31 March, 2003; originally announced March 2003.

New geometries associated with the nonlinear Schrödinger equation

Authors: S. Murugesh, Radha Balakrishnan

Abstract: We apply our recent formalism establishing new connections between the geometry of moving space curves and soliton equations, to the nonlinear Schrödinger equation (NLS). We show that any given solution of the NLS gets associated with three distinct space curve evolutions. The tangent vector of the first of these curves, the binormal vector of the second and the normal vector of the third, are… ▽ More We apply our recent formalism establishing new connections between the geometry of moving space curves and soliton equations, to the nonlinear Schrödinger equation (NLS). We show that any given solution of the NLS gets associated with three distinct space curve evolutions. The tangent vector of the first of these curves, the binormal vector of the second and the normal vector of the third, are shown to satisfy the integrable Landau-Lifshitz (LL) equation ${\bf S}_u = {\bf S} \times {\bf S}_{ss}$, (${\bf S}^2=1$). These connections enable us to find the three surfaces swept out by the moving curves associated with the NLS. As an example, surfaces corresponding to a stationary envelope soliton solution of the NLS are obtained. △ Less

Submitted 23 November, 2001; originally announced November 2001.

Comments: 13 pages, 3 figures

Kinematics of the Three Moving Space Curves Associated with the Nonlinear Schrödinger Equation

Authors: Radha Balakrishnan, S. Murugesh

Abstract: Starting with the general description of a moving curve, we have recently presented a unified formalism to show that three distinct space curve evolutions can be identified with a given integrable equation. Applying this to the nonlinear Schrödinger equation (NLS), we find three sets of coupled equations for the evolution of the curvature and the torsion, one set for each moving curve. The first… ▽ More Starting with the general description of a moving curve, we have recently presented a unified formalism to show that three distinct space curve evolutions can be identified with a given integrable equation. Applying this to the nonlinear Schrödinger equation (NLS), we find three sets of coupled equations for the evolution of the curvature and the torsion, one set for each moving curve. The first set is given by the usual Da Rios-Betchov equations. The velocity at each point of the curve that corresponds to this set is well known to be a local expression in the curve variables. In contrast, the velocities of the other two curves are shown to be nonlocal expressions. Each of the three curves is endowed with a corresponding infinite set of geometric constraints. These moving space curves are found by using their connection with the integrable Landau- Lifshitz equation. The three evolving curves corresponding to the envelope soliton solution of the NLS are presented and their behaviors compared. △ Less

Submitted 21 November, 2001; originally announced November 2001.

Comments: 15 pages, 3 figures, presented at NEEDS 2001, The Isaac Newton Institute of Mathematical Sciences, Cambridge, UK

New connections between moving curves and soliton equations

Authors: S. Murugesh, Radha Balakrishnan

Abstract: Lamb has identified a certain class of moving space curves with soliton equations. We show that there are two other classes of curve evolution that may be so identified. Hence three distinct classes of curve evolution are associated with a given integrable equation. The nonlinear Schrödinger equation is used to illustrate this explicitly. Lamb has identified a certain class of moving space curves with soliton equations. We show that there are two other classes of curve evolution that may be so identified. Hence three distinct classes of curve evolution are associated with a given integrable equation. The nonlinear Schrödinger equation is used to illustrate this explicitly. △ Less

Submitted 3 October, 2001; v1 submitted 27 April, 2001; originally announced April 2001.

Comments: some minor corrections made. 17 pages. Accepted in Phys. Lett. A

Nonlinear dynamics of the classical isotropic Heisenberg antiferromagnetic chain: the sigma model sector and the kink sector

Authors: Radha Balakrishnan, Rossen Dandoloff

Abstract: We identify two distinct low-energy sectors in the classical isotropic antiferromagnetic Heisenberg spin-S chain. In the continuum limit, we show that two types of rotation generators arise for the field in each sector. Using these, the Lagrangian for sector I is shown to be that of the nonlinear sigma model. Sector II has a null Lagrangian; Its Hamiltonian density is just the Pontryagin term. E… ▽ More We identify two distinct low-energy sectors in the classical isotropic antiferromagnetic Heisenberg spin-S chain. In the continuum limit, we show that two types of rotation generators arise for the field in each sector. Using these, the Lagrangian for sector I is shown to be that of the nonlinear sigma model. Sector II has a null Lagrangian; Its Hamiltonian density is just the Pontryagin term. Exact solutions are found in the form of magnons and precessing pulses in I and moving kinks in II. The kink has `spin' S. Sector I has a higher minimum energy than II. △ Less

Submitted 5 April, 2001; originally announced April 2001.

Comments: 4 pages