-
SMILe: Leveraging Submodular Mutual Information For Robust Few-Shot Object Detection
Authors:
Anay Majee,
Ryan Sharp,
Rishabh Iyer
Abstract:
Confusion and forgetting of object classes have been challenges of prime interest in Few-Shot Object Detection (FSOD). To overcome these pitfalls in metric learning based FSOD techniques, we introduce a novel Submodular Mutual Information Learning (SMILe) framework which adopts combinatorial mutual information functions to enforce the creation of tighter and discriminative feature clusters in FSOD…
▽ More
Confusion and forgetting of object classes have been challenges of prime interest in Few-Shot Object Detection (FSOD). To overcome these pitfalls in metric learning based FSOD techniques, we introduce a novel Submodular Mutual Information Learning (SMILe) framework which adopts combinatorial mutual information functions to enforce the creation of tighter and discriminative feature clusters in FSOD. Our proposed approach generalizes to several existing approaches in FSOD, agnostic of the backbone architecture demonstrating elevated performance gains. A paradigm shift from instance based objective functions to combinatorial objectives in SMILe naturally preserves the diversity within an object class resulting in reduced forgetting when subjected to few training examples. Furthermore, the application of mutual information between the already learnt (base) and newly added (novel) objects ensures sufficient separation between base and novel classes, minimizing the effect of class confusion. Experiments on popular FSOD benchmarks, PASCAL-VOC and MS-COCO show that our approach generalizes to State-of-the-Art (SoTA) approaches improving their novel class performance by up to 5.7% (3.3 mAP points) and 5.4% (2.6 mAP points) on the 10-shot setting of VOC (split 3) and 30-shot setting of COCO datasets respectively. Our experiments also demonstrate better retention of base class performance and up to 2x faster convergence over existing approaches agnostic of the underlying architecture.
△ Less
Submitted 2 July, 2024;
originally announced July 2024.
-
DeepOps & SLURM: Your GPU Cluster Guide
Authors:
Arindam Majee
Abstract:
In the ever evolving landscape of deep learning, unlocking the potential of cutting-edge models demands computational resources that surpass the capabilities of individual machines. Enter the NVIDIA DeepOps Slurm cluster, a meticulously orchestrated symphony of high-performance nodes, each equipped with powerful GPUs and meticulously managed by the efficient Slurm resource allocation system. This…
▽ More
In the ever evolving landscape of deep learning, unlocking the potential of cutting-edge models demands computational resources that surpass the capabilities of individual machines. Enter the NVIDIA DeepOps Slurm cluster, a meticulously orchestrated symphony of high-performance nodes, each equipped with powerful GPUs and meticulously managed by the efficient Slurm resource allocation system. This guide serves as your comprehensive roadmap, empowering you to harness the immense parallel processing capabilities of this cluster and propel your deep learning endeavors to new heights. Whether you are a seasoned deep learning practitioner seeking to optimize performance or a newcomer eager to unlock the power of parallel processing, this guide caters to your needs. We wll delve into the intricacies of the cluster hardware architecture, exploring the capabilities of its GPUs and the underlying network fabric. You will master the art of leveraging DeepOps containers for efficient and reproducible workflows, fine-tune resource configurations for optimal performance, and confidently submit jobs to unleash the full potential of parallel processing.
△ Less
Submitted 29 February, 2024;
originally announced May 2024.
-
Swarm UAVs Communication
Authors:
Arindam Majee,
Rahul Saha,
Snehasish Roy,
Srilekha Mandal,
Sayan Chatterjee
Abstract:
The advancement in cyber-physical systems has opened a new way in disaster management and rescue operations. The usage of UAVs is very promising in this context. UAVs, mainly quadcopters, are small in size and their payload capacity is limited. A single UAV can not traverse the whole area. Hence multiple UAVs or swarms of UAVs come into the picture managing the entire payload in a modular and equi…
▽ More
The advancement in cyber-physical systems has opened a new way in disaster management and rescue operations. The usage of UAVs is very promising in this context. UAVs, mainly quadcopters, are small in size and their payload capacity is limited. A single UAV can not traverse the whole area. Hence multiple UAVs or swarms of UAVs come into the picture managing the entire payload in a modular and equiproportional manner. In this work we have explored a vast topic related to UAVs. Among the UAVs quadcopter is the main focus. We explored the types of quadcopters, their flying strategy,their communication protocols, architecture and controlling techniques, followed by the swarm behaviour in nature and UAVs. Swarm behaviour and a few swarm optimization algorithms has been explored here. Swarm architecture and communication in between swarm UAV networks also got a special attention in our work. In disaster management the UAV swarm network must have to search a large area. And for this proper path planning algorithm is required. We have discussed the existing path planning algorithm, their advantages and disadvantages in great detail. Formation maintenance of the swarm network is an important issue which has been explored through leader-follower technique. The wireless path loss model has been modelled using friis and ground ray reflection model. Using this path loss models we have managed to create the link budget and simulate the variation of communication link performance with the variation of distance.
△ Less
Submitted 24 February, 2024;
originally announced May 2024.
-
Enhancing MRI-Based Classification of Alzheimer's Disease with Explainable 3D Hybrid Compact Convolutional Transformers
Authors:
Arindam Majee,
Avisek Gupta,
Sourav Raha,
Swagatam Das
Abstract:
Alzheimer's disease (AD), characterized by progressive cognitive decline and memory loss, presents a formidable global health challenge, underscoring the critical importance of early and precise diagnosis for timely interventions and enhanced patient outcomes. While MRI scans provide valuable insights into brain structures, traditional analysis methods often struggle to discern intricate 3D patter…
▽ More
Alzheimer's disease (AD), characterized by progressive cognitive decline and memory loss, presents a formidable global health challenge, underscoring the critical importance of early and precise diagnosis for timely interventions and enhanced patient outcomes. While MRI scans provide valuable insights into brain structures, traditional analysis methods often struggle to discern intricate 3D patterns crucial for AD identification. Addressing this challenge, we introduce an alternative end-to-end deep learning model, the 3D Hybrid Compact Convolutional Transformers 3D (HCCT). By synergistically combining convolutional neural networks (CNNs) and vision transformers (ViTs), the 3D HCCT adeptly captures both local features and long-range relationships within 3D MRI scans. Extensive evaluations on prominent AD benchmark dataset, ADNI, demonstrate the 3D HCCT's superior performance, surpassing state of the art CNN and transformer-based methods in classification accuracy. Its robust generalization capability and interpretability marks a significant stride in AD classification from 3D MRI scans, promising more accurate and reliable diagnoses for improved patient care and superior clinical outcomes.
△ Less
Submitted 24 March, 2024;
originally announced March 2024.
-
Charge Separation at Interfaces of Phase-Separated Coacervates
Authors:
Arghya Majee,
Christoph A. Weber,
Frank Jülicher
Abstract:
We present a theory for phase-separated liquid coacervates with salt, taking into account spatial heterogeneities and interfacial profiles. We find that charged layers of alternating sign can form around the interface while the bulk phases remain approximately charge-neutral. We show that the salt concentration regulates the number of layers and the amplitude of the layer's charge density and elec…
▽ More
We present a theory for phase-separated liquid coacervates with salt, taking into account spatial heterogeneities and interfacial profiles. We find that charged layers of alternating sign can form around the interface while the bulk phases remain approximately charge-neutral. We show that the salt concentration regulates the number of layers and the amplitude of the layer's charge density and electrostatic potential. Such charged layers can either repel or attract single-charged molecules diffusing across the interface. Our theory could be relevant for artificial systems and biomolecular condensates in cells. Our work suggests that interfaces of biomolecular condensates could mediate charge-specific transport similar to membrane-bound compartments.
△ Less
Submitted 11 October, 2023;
originally announced October 2023.
-
SCoRe: Submodular Combinatorial Representation Learning
Authors:
Anay Majee,
Suraj Kothawade,
Krishnateja Killamsetty,
Rishabh Iyer
Abstract:
In this paper we introduce the SCoRe (Submodular Combinatorial Representation Learning) framework, a novel approach in representation learning that addresses inter-class bias and intra-class variance. SCoRe provides a new combinatorial viewpoint to representation learning, by introducing a family of loss functions based on set-based submodular information measures. We develop two novel combinatori…
▽ More
In this paper we introduce the SCoRe (Submodular Combinatorial Representation Learning) framework, a novel approach in representation learning that addresses inter-class bias and intra-class variance. SCoRe provides a new combinatorial viewpoint to representation learning, by introducing a family of loss functions based on set-based submodular information measures. We develop two novel combinatorial formulations for loss functions, using the Total Information and Total Correlation, that naturally minimize intra-class variance and inter-class bias. Several commonly used metric/contrastive learning loss functions like supervised contrastive loss, orthogonal projection loss, and N-pairs loss, are all instances of SCoRe, thereby underlining the versatility and applicability of SCoRe in a broad spectrum of learning scenarios. Novel objectives in SCoRe naturally model class-imbalance with up to 7.6\% improvement in classification on CIFAR-10-LT, CIFAR-100-LT, MedMNIST, 2.1% on ImageNet-LT, and 19.4% in object detection on IDD and LVIS (v1.0), demonstrating its effectiveness over existing approaches.
△ Less
Submitted 6 June, 2024; v1 submitted 29 September, 2023;
originally announced October 2023.
-
Electrostatic interactions between charge regulated spherical macroions
Authors:
Hu Ruixuan,
Arghya Majee,
Jure Dobnikar,
Rudolf Podgornik
Abstract:
We study the interaction between two charge regulating spherical macroions with dielectric interior and dissociable surface groups immersed in a monovalent electrolyte solution. The charge dissociation is modelled via the Frumkin-Fowler-Guggenheim isotherm, which allows for multiple adsorption equilibrium states. The interactions are derived from the solutions of the mean-field Poisson-Boltzmann t…
▽ More
We study the interaction between two charge regulating spherical macroions with dielectric interior and dissociable surface groups immersed in a monovalent electrolyte solution. The charge dissociation is modelled via the Frumkin-Fowler-Guggenheim isotherm, which allows for multiple adsorption equilibrium states. The interactions are derived from the solutions of the mean-field Poisson-Boltzmann type theory with charge regulation boundary conditions. For a range of conditions we find symmetry breaking transitions from symmetric to asymmetric charge distribution exhibiting annealed charge patchiness, which results in like-charge attraction even in a univalent electrolyte -- thus fundamentally modifying the nature of electrostatic interactions in charge-stabilized colloidal suspensions.
△ Less
Submitted 7 December, 2023; v1 submitted 6 August, 2023;
originally announced August 2023.
-
Nonlinear SPDE driven by Levy noise: Well-posedness, optimal control and invariant measure
Authors:
Kavin R,
Ananta K. Majee
Abstract:
In this article, we study a nonlinear stochastic control problem perturbed by multiplicative Levy noise, where the nonlinear operator in divergence form satisfies p type growth with coercivity assumptions. By using Aldous tightness criteria and Jakubowski version of the Skorokhod theorem on nonmetric spaces along with standard contraction method, we establish existence of pathwise unique strong so…
▽ More
In this article, we study a nonlinear stochastic control problem perturbed by multiplicative Levy noise, where the nonlinear operator in divergence form satisfies p type growth with coercivity assumptions. By using Aldous tightness criteria and Jakubowski version of the Skorokhod theorem on nonmetric spaces along with standard contraction method, we establish existence of pathwise unique strong solution. Formulating the associated control problem, and using variational approach together with the convexity property of cost functional in control variable, we establish existence of a weak optimal solution of the underlying problem. We use the technique of Maslowski and Seidler to prove existence of an invariant measure for uncontrolled SPDE driven with multiplicative Levy noise.
△ Less
Submitted 7 June, 2023;
originally announced June 2023.
-
Stochastic Fractional Conservation Laws: Large deviation principle, Central limit theorem and Moderate deviation principle
Authors:
Soumya Ranjan Behera,
Ananta K. Majee
Abstract:
In this article, we establish the Freidlin-Wentzell type large deviation principle and central limit theorem for stochastic fractional conservation laws with small multiplicative noise in kinetic formulation framework. The weak convergence method and doubling variables method play a crucial role. As a consequence, we also establish moderate deviation principle for the underlying problem.
In this article, we establish the Freidlin-Wentzell type large deviation principle and central limit theorem for stochastic fractional conservation laws with small multiplicative noise in kinetic formulation framework. The weak convergence method and doubling variables method play a crucial role. As a consequence, we also establish moderate deviation principle for the underlying problem.
△ Less
Submitted 7 June, 2023;
originally announced June 2023.
-
A Comparative Analysis of Portfolio Optimization Using Mean-Variance, Hierarchical Risk Parity, and Reinforcement Learning Approaches on the Indian Stock Market
Authors:
Jaydip Sen,
Aditya Jaiswal,
Anshuman Pathak,
Atish Kumar Majee,
Kushagra Kumar,
Manas Kumar Sarkar,
Soubhik Maji
Abstract:
This paper presents a comparative analysis of the performances of three portfolio optimization approaches. Three approaches of portfolio optimization that are considered in this work are the mean-variance portfolio (MVP), hierarchical risk parity (HRP) portfolio, and reinforcement learning-based portfolio. The portfolios are trained and tested over several stock data and their performances are com…
▽ More
This paper presents a comparative analysis of the performances of three portfolio optimization approaches. Three approaches of portfolio optimization that are considered in this work are the mean-variance portfolio (MVP), hierarchical risk parity (HRP) portfolio, and reinforcement learning-based portfolio. The portfolios are trained and tested over several stock data and their performances are compared on their annual returns, annual risks, and Sharpe ratios. In the reinforcement learning-based portfolio design approach, the deep Q learning technique has been utilized. Due to the large number of possible states, the construction of the Q-table is done using a deep neural network. The historical prices of the 50 premier stocks from the Indian stock market, known as the NIFTY50 stocks, and several stocks from 10 important sectors of the Indian stock market are used to create the environment for training the agent.
△ Less
Submitted 27 May, 2023;
originally announced May 2023.
-
Curvature effects in interfacial acidity of amphiphilic vesicles
Authors:
Petch Khunpetch,
Arghya Majee,
Hu Ruixuan,
Rudolf Podgornik
Abstract:
We analyze the changes in the vicinal acidity (pH) at a spherical amphiphilic membrane. The membrane is assumed to contain solvent accessible, embedded, dissociable, charge regulated moieties. Basing our approach on the linear Debye-Hückel as well as the non-linear Poisson-Boltzmann theory, together with the general Frumkin-Fowler-Guggenheim adsorption isotherm model of the charge regulation proce…
▽ More
We analyze the changes in the vicinal acidity (pH) at a spherical amphiphilic membrane. The membrane is assumed to contain solvent accessible, embedded, dissociable, charge regulated moieties. Basing our approach on the linear Debye-Hückel as well as the non-linear Poisson-Boltzmann theory, together with the general Frumkin-Fowler-Guggenheim adsorption isotherm model of the charge regulation process, we analyse and review the dependence of the local pH on the position, as well as bulk electrolyte concentration, bulk pH and curvature of the amphiphilic single membrane vesicle. With appropriately chosen adsorption parameters of the charge regulation model, we find a good agreement with available experimental data.
△ Less
Submitted 18 May, 2023;
originally announced May 2023.
-
On rate of convergence of finite difference scheme for degenerate parabolic-hyperbolic PDE with Levy noise
Authors:
Soumya Ranjan Behera,
Ananta K. Majee
Abstract:
In this article, we consider a semi discrete finite difference scheme for a degenerate parabolic-hyperbolic PDE driven by Lévy noise in one space dimension. Using bounded variation estimations and a variant of classical Kružkov's doubling of variable approach, we prove that expected value of the $L^1$-difference between the unique entropy solution and approximate solution converges at a rate of…
▽ More
In this article, we consider a semi discrete finite difference scheme for a degenerate parabolic-hyperbolic PDE driven by Lévy noise in one space dimension. Using bounded variation estimations and a variant of classical Kružkov's doubling of variable approach, we prove that expected value of the $L^1$-difference between the unique entropy solution and approximate solution converges at a rate of $(Δx)^\frac{1}{7}$, where $Δx$ is the spatial mesh size.
△ Less
Submitted 20 December, 2023; v1 submitted 24 December, 2022;
originally announced December 2022.
-
Stochastic doubly nonlinear PDE: Large Deviation Principles and existence of Invariant measure
Authors:
Ananta K Majee
Abstract:
In this paper, we establish large deviation principle for the strong solution of a doubly nonlinear PDE driven by small multiplicative Brownian noise. Motononicity arguments and the weak convergence approach have been exploited in the proof. Moreover, by using certain a-priori estimates and sequentially weakly Feller property of the associated Markov semigroup, we show existence of invariant proba…
▽ More
In this paper, we establish large deviation principle for the strong solution of a doubly nonlinear PDE driven by small multiplicative Brownian noise. Motononicity arguments and the weak convergence approach have been exploited in the proof. Moreover, by using certain a-priori estimates and sequentially weakly Feller property of the associated Markov semigroup, we show existence of invariant probability measure for the strong solution of the underlying problem.
△ Less
Submitted 24 December, 2022;
originally announced December 2022.
-
Convergence of an operator splitting scheme for fractional conservation laws with Levy noise
Authors:
Soumya Ranjan Behera,
Ananta K. Majee
Abstract:
In this paper, we are concerned with a operator splitting scheme for linear fractional and fractional degenerate stochastic conservation laws driven by multiplicative Levy noise. More specifically, using a variant of classical Kruzkov's doubling of variable approach, we show that the approximate solutions generated by the splitting scheme converges to the unique stochastic entropy solution of the…
▽ More
In this paper, we are concerned with a operator splitting scheme for linear fractional and fractional degenerate stochastic conservation laws driven by multiplicative Levy noise. More specifically, using a variant of classical Kruzkov's doubling of variable approach, we show that the approximate solutions generated by the splitting scheme converges to the unique stochastic entropy solution of the underlying problems.Finally, the convergence analysis is illustrated by several numerical examples.
△ Less
Submitted 11 March, 2023; v1 submitted 5 December, 2022;
originally announced December 2022.
-
Stochastic evolutionary p-Laplace equation: Large Deviation Principles and Transportation Cost Inequality
Authors:
Kavin R,
Ananta K Majee
Abstract:
In this paper, we establish large deviation principle for the strong solution of evolutionary p-Laplace equation driven by small multiplicative Brownian noise, where the weak convergence approach plays a key role. Moreover, by using Girsanov transformation along with $L^1$-contraction approach, we show the quadratic transportation cost inequality for the strong solution of the underlying problem.
In this paper, we establish large deviation principle for the strong solution of evolutionary p-Laplace equation driven by small multiplicative Brownian noise, where the weak convergence approach plays a key role. Moreover, by using Girsanov transformation along with $L^1$-contraction approach, we show the quadratic transportation cost inequality for the strong solution of the underlying problem.
△ Less
Submitted 20 October, 2022;
originally announced October 2022.
-
Curvature effects in charge-regulated lipid bilayers
Authors:
Petch Khunpetch,
Arghya Majee,
Rudolf Podgornik
Abstract:
We formulate a theory of electrostatic interactions in lipid bilayer membranes where both monolayer leaflets contain dissociable moieties that are subject to charge regulation. We specifically investigate the coupling between membrane curvature and charge regulation of a lipid bilayer vesicle using both the linear Debye-Hückel (DH) and the non-linear Poisson-Boltzmann (PB) theory. We find that cha…
▽ More
We formulate a theory of electrostatic interactions in lipid bilayer membranes where both monolayer leaflets contain dissociable moieties that are subject to charge regulation. We specifically investigate the coupling between membrane curvature and charge regulation of a lipid bilayer vesicle using both the linear Debye-Hückel (DH) and the non-linear Poisson-Boltzmann (PB) theory. We find that charge regulation of an otherwise symmetric bilayer membrane can induce charge symmetry breaking, non-linear flexoelectricity and anomalous curvature dependence of free energy. The pH effects investigated go beyond the paradigm of electrostatic renormalization of the mechano-elastic properties of membranes.
△ Less
Submitted 15 February, 2022; v1 submitted 13 January, 2022;
originally announced January 2022.
-
Electroviscous drag on squeezing motion in sphere-plane geometry
Authors:
Marcela Rodriguez Matus,
Zaicheng Zhang,
Zouhir Benrahla,
Arghya Majee,
Abdelhamid Maali,
Alois Würger
Abstract:
Theoretically and experimentally, we study electroviscous phenomena resulting from charge-flow coupling in a nanoscale capillary. Our theoretical approach relies on Poisson-Boltzmann mean-field theory and on coupled linear relations for charge and hydrodynamic flows, including electro-osmosis and charge advection. With respect to the unperturbed Poiseuille flow, we define an electroviscous couplin…
▽ More
Theoretically and experimentally, we study electroviscous phenomena resulting from charge-flow coupling in a nanoscale capillary. Our theoretical approach relies on Poisson-Boltzmann mean-field theory and on coupled linear relations for charge and hydrodynamic flows, including electro-osmosis and charge advection. With respect to the unperturbed Poiseuille flow, we define an electroviscous coupling parameter $ξ$, which turns out to be maximum where the film thickness $h_0$ is comparable to the screening length $λ$. We also present dynamic AFM data for the visco-elastic response of a confined water film in sphere-plane geometry; our theory provides a quantitative description for the electroviscous drag coefficient and the electrostatic repulsion as a function of the film thickness, with the surface charge density as the only free parameter. Charge regulation sets in at even smaller distances.
△ Less
Submitted 28 May, 2022; v1 submitted 4 January, 2022;
originally announced January 2022.
-
Attention Guided Cosine Margin For Overcoming Class-Imbalance in Few-Shot Road Object Detection
Authors:
Ashutosh Agarwal,
Anay Majee,
Anbumani Subramanian,
Chetan Arora
Abstract:
Few-shot object detection (FSOD) localizes and classifies objects in an image given only a few data samples. Recent trends in FSOD research show the adoption of metric and meta-learning techniques, which are prone to catastrophic forgetting and class confusion. To overcome these pitfalls in metric learning based FSOD techniques, we introduce Attention Guided Cosine Margin (AGCM) that facilitates t…
▽ More
Few-shot object detection (FSOD) localizes and classifies objects in an image given only a few data samples. Recent trends in FSOD research show the adoption of metric and meta-learning techniques, which are prone to catastrophic forgetting and class confusion. To overcome these pitfalls in metric learning based FSOD techniques, we introduce Attention Guided Cosine Margin (AGCM) that facilitates the creation of tighter and well separated class-specific feature clusters in the classification head of the object detector. Our novel Attentive Proposal Fusion (APF) module minimizes catastrophic forgetting by reducing the intra-class variance among co-occurring classes. At the same time, the proposed Cosine Margin Cross-Entropy loss increases the angular margin between confusing classes to overcome the challenge of class confusion between already learned (base) and newly added (novel) classes. We conduct our experiments on the challenging India Driving Dataset (IDD), which presents a real-world class-imbalanced setting alongside popular FSOD benchmark PASCAL-VOC. Our method outperforms State-of-the-Art (SoTA) approaches by up to 6.4 mAP points on the IDD-OS and up to 2.0 mAP points on the IDD-10 splits for the 10-shot setting. On the PASCAL-VOC dataset, we outperform existing SoTA approaches by up to 4.9 mAP points.
△ Less
Submitted 12 November, 2021;
originally announced November 2021.
-
Meta Guided Metric Learner for Overcoming Class Confusion in Few-Shot Road Object Detection
Authors:
Anay Majee,
Anbumani Subramanian,
Kshitij Agrawal
Abstract:
Localization and recognition of less-occurring road objects have been a challenge in autonomous driving applications due to the scarcity of data samples. Few-Shot Object Detection techniques extend the knowledge from existing base object classes to learn novel road objects given few training examples. Popular techniques in FSOD adopt either meta or metric learning techniques which are prone to cla…
▽ More
Localization and recognition of less-occurring road objects have been a challenge in autonomous driving applications due to the scarcity of data samples. Few-Shot Object Detection techniques extend the knowledge from existing base object classes to learn novel road objects given few training examples. Popular techniques in FSOD adopt either meta or metric learning techniques which are prone to class confusion and base class forgetting. In this work, we introduce a novel Meta Guided Metric Learner (MGML) to overcome class confusion in FSOD. We re-weight the features of the novel classes higher than the base classes through a novel Squeeze and Excite module and encourage the learning of truly discriminative class-specific features by applying an Orthogonality Constraint to the meta learner. Our method outperforms State-of-the-Art (SoTA) approaches in FSOD on the India Driving Dataset (IDD) by upto 11 mAP points while suffering from the least class confusion of 20% given only 10 examples of each novel road object. We further show similar improvements on the few-shot splits of PASCAL VOC dataset where we outperform SoTA approaches by upto 5.8 mAP accross all splits.
△ Less
Submitted 28 October, 2021;
originally announced October 2021.
-
Few-Shot Batch Incremental Road Object Detection via Detector Fusion
Authors:
Anuj Tambwekar,
Kshitij Agrawal,
Anay Majee,
Anbumani Subramanian
Abstract:
Incremental few-shot learning has emerged as a new and challenging area in deep learning, whose objective is to train deep learning models using very few samples of new class data, and none of the old class data. In this work we tackle the problem of batch incremental few-shot road object detection using data from the India Driving Dataset (IDD). Our approach, DualFusion, combines object detectors…
▽ More
Incremental few-shot learning has emerged as a new and challenging area in deep learning, whose objective is to train deep learning models using very few samples of new class data, and none of the old class data. In this work we tackle the problem of batch incremental few-shot road object detection using data from the India Driving Dataset (IDD). Our approach, DualFusion, combines object detectors in a manner that allows us to learn to detect rare objects with very limited data, all without severely degrading the performance of the detector on the abundant classes. In the IDD OpenSet incremental few-shot detection task, we achieve a mAP50 score of 40.0 on the base classes and an overall mAP50 score of 38.8, both of which are the highest to date. In the COCO batch incremental few-shot detection task, we achieve a novel AP score of 9.9, surpassing the state-of-the-art novel class performance on the same by over 6.6 times.
△ Less
Submitted 18 August, 2021;
originally announced August 2021.
-
Few-Shot Learning for Road Object Detection
Authors:
Anay Majee,
Kshitij Agrawal,
Anbumani Subramanian
Abstract:
Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of less-occurring road objects in the image dataset and hence provides a setup suitable for few-shot learning.…
▽ More
Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of less-occurring road objects in the image dataset and hence provides a setup suitable for few-shot learning. We evaluate both metric-learning and meta-learning based FSOD methods, in two experimental settings: (i) representative (same-domain) splits from IDD, that evaluates the ability of a model to learn in the context of road images, and (ii) object classes with less-occurring object samples, similar to the open-set setting in real-world. From our experiments, we demonstrate that the metric-learning method outperforms meta-learning on the novel classes by (i) 11.2 mAP points on the same domain, and (ii) 1.0 mAP point on the open-set. We also show that our extension of object classes in a real-world open dataset offers a rich ground for few-shot learning studies.
△ Less
Submitted 17 March, 2021; v1 submitted 29 January, 2021;
originally announced January 2021.
-
Charge symmetry broken complex coacervation
Authors:
Arghya Majee,
Markus Bier,
Ralf Blossey,
Rudolf Podgornik
Abstract:
Liquid-liquid phase separation has emerged as one of the important paradigms in the chemical physics as well as biophysics of charged macromolecular systems. We elucidate an equilibrium phase separation mechanism based on charge regulation, i.e., protonation-deprotonation equilibria controlled by pH, in an idealized macroion system which can serve as a proxy for simple coacervation. First, a low-d…
▽ More
Liquid-liquid phase separation has emerged as one of the important paradigms in the chemical physics as well as biophysics of charged macromolecular systems. We elucidate an equilibrium phase separation mechanism based on charge regulation, i.e., protonation-deprotonation equilibria controlled by pH, in an idealized macroion system which can serve as a proxy for simple coacervation. First, a low-density density-functional calculation reveals the dominance of two-particle configurations coupled by ion adsorption on neighboring macroions. Then a binary cell model, solved on the Debye-Hückel as well as the full nonlinear Poisson-Boltzmann level, unveils the charge-symmetry breaking as inducing the phase separation between low- and high-density phases as a function of pH. These results can be identified as a charge symmetry broken complex coacervation between chemically identical macroions.
△ Less
Submitted 24 December, 2020; v1 submitted 1 October, 2020;
originally announced October 2020.
-
Electrostatic pair-interaction of nearby metal or metal-coated colloids at fluid interfaces
Authors:
Rick Bebon,
Arghya Majee
Abstract:
In this paper, we theoretically study the electrostatic interaction between a pair of identical colloids with constant surface potentials sitting in close vicinity of each other at a fluid interface. By employing a simplified yet reasonable model system, the problem is solved within the framework of classical density functional theory and linearized as well as nonlinear Poisson-Boltzmann (PB) theo…
▽ More
In this paper, we theoretically study the electrostatic interaction between a pair of identical colloids with constant surface potentials sitting in close vicinity of each other at a fluid interface. By employing a simplified yet reasonable model system, the problem is solved within the framework of classical density functional theory and linearized as well as nonlinear Poisson-Boltzmann (PB) theory. Apart from providing a sound theoretical framework generally applicable to any such problem, our novel findings, all of which contradict common beliefs, include the following: first, quantitative as well as qualitative differences between the interactions obtained within the linear and the nonlinear PB theories; second, the importance of the electrostatic interaction between the omnipresent three-phase contact lines in interfacial systems; and third, the occurrence of an attractive electrostatic interaction between a pair of identical metal colloids. The unusual attraction we report on largely stems from an attractive line interaction which although scales linearly with the size of the particle, can compete with the surface interactions and can be strong enough to alter the nature of the total electrostatic interaction. Our results should find applications in metal or metal-coated particle-stabilized emulsions where densely packed particle arrays are not only frequently observed but are sometimes required.
△ Less
Submitted 28 July, 2020;
originally announced July 2020.
-
Charge regulation radically modifies electrostatics in membrane stacks
Authors:
Arghya Majee,
Markus Bier,
Ralf Blossey,
Rudolf Podgornik
Abstract:
Motivated by biological membrane-containing organelles in plants and photosynthetic bacteria, we study charge regulation in a model membrane stack. Considering (de)protonation as the simplest mechanism of charge equilibration between the membranes and with the bathing environment, we uncover a symmetry-broken charge state in the stack with a quasiperiodic effective charge sequence. In the case of…
▽ More
Motivated by biological membrane-containing organelles in plants and photosynthetic bacteria, we study charge regulation in a model membrane stack. Considering (de)protonation as the simplest mechanism of charge equilibration between the membranes and with the bathing environment, we uncover a symmetry-broken charge state in the stack with a quasiperiodic effective charge sequence. In the case of a monovalent bathing salt solution our model predicts complex, inhomogeneous charge equilibria depending on the strength of the (de)protonation reaction, salt concentration, intermembrane separation, and their number in the stack. Our results shed light on the basic reorganization mechanism of thylakoid membrane stacks.
△ Less
Submitted 6 November, 2019; v1 submitted 31 August, 2019;
originally announced September 2019.
-
On the Development of a Coupled Non-linear Telegraph-Diffusion Model for Image Restoration
Authors:
Sudeb Majee,
Subit K. Jain,
Rajendra K. Ray,
Ananta K. Majee
Abstract:
In this work, we propose a telegraph coupled partial differential equation (TCPDE) based model for image restoration. New framework interpolates between a couple of non-linear telegraph equation and a parabolic equation. Proposed strategy can be applied to significantly preserve the oscillatory and texture pattern in an image, even in low signal-to-noise ratio. First, we prove that the present mod…
▽ More
In this work, we propose a telegraph coupled partial differential equation (TCPDE) based model for image restoration. New framework interpolates between a couple of non-linear telegraph equation and a parabolic equation. Proposed strategy can be applied to significantly preserve the oscillatory and texture pattern in an image, even in low signal-to-noise ratio. First, we prove that the present model has a unique global weak solution using Banach's fixed point theorem. Then apply our model over a set of gray-level images to illustrate the superiority of the proposed model over the recently developed hyperbolic-parabolic PDE based models as well as coupled diffusion-based model.
△ Less
Submitted 11 September, 2019; v1 submitted 30 August, 2019;
originally announced August 2019.
-
Well-posedness study of a non-linear hyperbolic-parabolic coupled system applied to image speckle reduction
Authors:
Sudeb Majee,
Rajendra K. Ray,
Ananta K. Majee
Abstract:
In this article, we consider a non-linear hyperbolic-parabolic coupled system based on telegraph diffusion framework applied to image despeckling. A separate equation is used to calculate the edge variable, which improves the quality of the despeckled images. A well-posedness result of the proposed coupled system is settled via Schauder's fixed point theorem. Numerical experiments are reported to…
▽ More
In this article, we consider a non-linear hyperbolic-parabolic coupled system based on telegraph diffusion framework applied to image despeckling. A separate equation is used to calculate the edge variable, which improves the quality of the despeckled images. A well-posedness result of the proposed coupled system is settled via Schauder's fixed point theorem. Numerical experiments are reported to illustrate the effectiveness of the proposed model, with recently developed models, over a set of gray level test images contaminated by speckle noise.
△ Less
Submitted 11 August, 2019; v1 submitted 7 August, 2019;
originally announced August 2019.
-
A Gray Level Indicator-Based Regularized Telegraph Diffusion Equation Applied to Image Despeckling
Authors:
Sudeb Majee,
Rajendra K Ray,
Ananta K Majee
Abstract:
In this work, a gray level indicator based non-linear telegraph diffusion model is presented for multiplicative noise removal problem. Most of the researchers focus only on diffusion equation-based model for multiplicative noise removal problem. The suggested model uses the benefit of the combined effect of diffusion equation as well as the wave equation. Wave nature of the model preserves the hig…
▽ More
In this work, a gray level indicator based non-linear telegraph diffusion model is presented for multiplicative noise removal problem. Most of the researchers focus only on diffusion equation-based model for multiplicative noise removal problem. The suggested model uses the benefit of the combined effect of diffusion equation as well as the wave equation. Wave nature of the model preserves the high oscillatory and texture pattern in an image. In this model, the diffusion coefficient depends not only on the image gradient but also on the gray level of the image, which controls the diffusion process better than only gradient-based diffusion models. Moreover, we prove the well-posedness of the present model using Schauder fixed point theorem. Furthermore, we show the superiority of the proposed model over a recently developed method on a set of gray level test images which are corrupted by speckle noise.
△ Less
Submitted 6 August, 2019; v1 submitted 3 August, 2019;
originally announced August 2019.
-
A Fuzzy Edge Detector Driven Telegraph Total Variation Model For Image Despeckling
Authors:
Sudeb Majee,
Subit K Jain,
Rajendra K Ray,
Ananta K Majee
Abstract:
Speckle noise suppression is a challenging and crucial pre-processing stage for higher-level image analysis. In this work, a new attempt has been made using telegraph total variation equation and fuzzy set theory for speckle noise suppression. The intuitionistic fuzzy divergence (IFD) function has been used to distinguish between edges and noise. To the best of the author's knowledge, most of the…
▽ More
Speckle noise suppression is a challenging and crucial pre-processing stage for higher-level image analysis. In this work, a new attempt has been made using telegraph total variation equation and fuzzy set theory for speckle noise suppression. The intuitionistic fuzzy divergence (IFD) function has been used to distinguish between edges and noise. To the best of the author's knowledge, most of the studies on multiplicative speckle noise removal process focus on only diffusion-based filters, and little attention has been paid to the study of fuzzy set theory. The proposed approach enjoy the benefits of both telegraph total variation equation and fuzzy edge detector, which is not only robust to noise but also preserves image structural details. Moreover, we establish the existence and uniqueness of a weak solution of the regularized version of the proposed model using Schauder fixed point theorem. With the proposed model, despeckling is carried out on natural and Synthetic Aperture Radar (SAR) images. The experimental results of the proposed model are reported, which found better in terms of noise suppression and detail/edge preservation, with respect to the existing approaches.
△ Less
Submitted 5 August, 2019; v1 submitted 3 August, 2019;
originally announced August 2019.
-
Analysis and Simulation of a Coupled Diffusion based Image Denoising Model
Authors:
Subit K. Jain,
Sudeb Majee,
Rajendra K. Ray,
Ananta K. Majee
Abstract:
In this study, a new coupled Partial Differential Equation (CPDE) based image denoising model incorporating space-time regularization into non-linear diffusion is proposed. This proposed model is fitted with additive Gaussian noise which performs efficient image smoothing along with the preservation of edges and fine structures. For this purpose, we propose a new functional minimization framework…
▽ More
In this study, a new coupled Partial Differential Equation (CPDE) based image denoising model incorporating space-time regularization into non-linear diffusion is proposed. This proposed model is fitted with additive Gaussian noise which performs efficient image smoothing along with the preservation of edges and fine structures. For this purpose, we propose a new functional minimization framework to remove the image noise, which results in solving a system of three partial differential equations (PDEs). Our proposed model is dissimilar from the existing CPDE models as it includes two additional evolution equations to handle edge strength function and data fidelity term. These two evolution equations control the smoothing process and force the resultant denoised solution to be close to the initial solution. To the best of our knowledge, the proposed model is the only work, which deciphers the combined effect of both the terms using separate PDEs. Furthermore, we establish the existence and uniqueness of a weak solution of the proposed system using the time discretization method with $H^1$ initial data. Finally, we used a generalized weighted average finite difference scheme to efficiently solve the coupled system and experiment results show the effectiveness of the proposed CPDE model.
△ Less
Submitted 7 August, 2019; v1 submitted 10 July, 2019;
originally announced July 2019.
-
Stochastic optimal control of a evolutionary $p$-Laplace equation with multiplicative Lévy noise
Authors:
Ananta K. Majee
Abstract:
In this article, we are interested in an initial value optimal control problem for a evolutionary $p$-Laplace equation driven by multiplicative Lévy noise. We first present wellposedness of a weak solution by using an implicit time discretization of the problem, along with the Jakubowski version of the Skorokhod theorem for a non-metric space. We then formulate associated control problem, and esta…
▽ More
In this article, we are interested in an initial value optimal control problem for a evolutionary $p$-Laplace equation driven by multiplicative Lévy noise. We first present wellposedness of a weak solution by using an implicit time discretization of the problem, along with the Jakubowski version of the Skorokhod theorem for a non-metric space. We then formulate associated control problem, and establish existence of an optimal solution by using variational method and exploiting the convexity property of the cost functional.
△ Less
Submitted 8 July, 2019;
originally announced July 2019.
-
Electronic Transport and Thermopower in 2D and 3D Heterostructures--A Theory Perspective
Authors:
Arnab K. Majee,
Adithya Kommini,
Zlatan Aksamija
Abstract:
In this review, we discuss the impact of interfaces and heterojuctions on the electronic and thermoelectric transport properties of materials. We review recent progress in understanding electronic transport in two-dimensional (2D) materials ranging from graphene to transition metal dichalcogenides (TMDs), their homojunctions (grain boundaries), lateral heterojunctions (such as graphene/MoS$_2$ lat…
▽ More
In this review, we discuss the impact of interfaces and heterojuctions on the electronic and thermoelectric transport properties of materials. We review recent progress in understanding electronic transport in two-dimensional (2D) materials ranging from graphene to transition metal dichalcogenides (TMDs), their homojunctions (grain boundaries), lateral heterojunctions (such as graphene/MoS$_2$ lateral interfaces), and vertical van der Waals (vdW) heterostructures. We also review work in thermoelectric properties of 2D heterojunctions, as well as their applications in creating devices such as resonant tunneling diodes (RTDs). Lastly, we turn our focus to work in three-dimensional (3D) heterostructures. While transport in 3D heterostructures has been researched for several decades, here we review recent progress in theory and simulation of quantum effects on transport via the Wigner and non-equilibrium Green's functions (NEGF) approaches. These simulation techniques have been successfully applied toward understanding the impact of heterojunctions on the thermoelectric properties, with applications in energy harvesting, and electron resonant tunneling, with applications in RTDs. We conclude that tremendous progress has been made in both simulation and experiments toward the goal of understanding transport in heterostructures and this progress will soon be parlayed into improved energy converters and quantum nanoelectronic devices.
△ Less
Submitted 14 March, 2019; v1 submitted 30 January, 2019;
originally announced January 2019.
-
Electrostatic interaction of particles trapped at fluid interfaces: Effects of geometry and wetting properties
Authors:
Arghya Majee,
Markus Bier,
S. Dietrich
Abstract:
The electrostatic interaction between pairs of spherical or macroscopically long, parallel cylindrical colloids trapped at fluid interfaces is studied theoretically for the case of small inter-particle separations. Starting from the effective interaction between two planar walls and by using the Derjaguin approximation, we address the issue of how the electrostatic interaction between such particl…
▽ More
The electrostatic interaction between pairs of spherical or macroscopically long, parallel cylindrical colloids trapped at fluid interfaces is studied theoretically for the case of small inter-particle separations. Starting from the effective interaction between two planar walls and by using the Derjaguin approximation, we address the issue of how the electrostatic interaction between such particles is influenced by their curvatures and by the wetting contact angle at their surfaces. Regarding the influence of curvature, our findings suggest that the discrepancies between linear and nonlinear Poisson-Boltzmann theory, which have been noticed before for planar walls, also occur for spheres and macroscopically long, parallel cylinders, though their magnitude depends on the wetting contact angle. Concerning the influence of the wetting contact angle $θ$ simple relations are obtained for equally sized particles which indicate that the inter-particle force varies significantly with $θ$ only within an interval around $90^\circ$. This interval depends on the Debye length of the fluids and on the size of the particles but not on their shape. For unequally sized particles, a more complicated relation is obtained for the variation of the inter-particle force with the wetting contact angle.
△ Less
Submitted 29 November, 2018; v1 submitted 29 August, 2018;
originally announced August 2018.
-
Dynamic Programming for Finite Ensembles of Nanomagnetic Particles
Authors:
Max Jensen,
Ananta Majee,
Andreas Prohl,
Christian Schellnegger
Abstract:
We use optimal control via a distributed exterior field to steer the dynamics of an ensemble of N interacting ferromagnetic particles which are immersed into a heat bath by minimizing a quadratic functional. By using dynamic programing principle, we show the existence of a unique strong solution of the optimal control problem. By the Hopf-Cole transformation, the related Hamilton-Jacobi-Bellman eq…
▽ More
We use optimal control via a distributed exterior field to steer the dynamics of an ensemble of N interacting ferromagnetic particles which are immersed into a heat bath by minimizing a quadratic functional. By using dynamic programing principle, we show the existence of a unique strong solution of the optimal control problem. By the Hopf-Cole transformation, the related Hamilton-Jacobi-Bellman equation from dynamic programming principle may be re-cast into a linear PDE on the manifold M = (S^2)^N, whose classical solution may be represented via Feynman-Kac formula. We use this probabilistic representation for Monte-Carlo simulations to illustrate optimal switching dynamics.
△ Less
Submitted 18 June, 2018;
originally announced June 2018.
-
Electrostatic interaction between dissimilar colloids at fluid interfaces
Authors:
Arghya Majee,
Timo Schmetzer,
Markus Bier
Abstract:
The electrostatic interaction between two non-identical, moderately charged colloids situated in close proximity of each other at a fluid interface is studied. By resorting to a well-justified model system, this problem is analytically solved within the framework of linearized Poisson-Boltzmann (PB) density functional theory (DFT). The resulting interaction comprises a surface and a line part, bot…
▽ More
The electrostatic interaction between two non-identical, moderately charged colloids situated in close proximity of each other at a fluid interface is studied. By resorting to a well-justified model system, this problem is analytically solved within the framework of linearized Poisson-Boltzmann (PB) density functional theory (DFT). The resulting interaction comprises a surface and a line part, both of which, as functions of the inter-particle separation, show a rich behavior including monotonic as well as non-monotonic variations. In almost all cases, these variations cannot be captured correctly by using the superposition approximation. Moreover, expressions for the surface tensions, the line tensions and the fluid-fluid interfacial tension, which are all independent of the inter-particle separation, are obtained. Our results are expected to be particularly useful for emulsions stabilized by oppositely charged particles.
△ Less
Submitted 27 April, 2018; v1 submitted 17 January, 2018;
originally announced January 2018.
-
Spontaneous symmetry breaking of charge-regulated surfaces
Authors:
Arghya Majee,
Markus Bier,
Rudolf Podgornik
Abstract:
The interaction between two chemically identical charge-regulated surfaces is studied using the classical density functional theory. In contrast to common expectations and assumptions, under certain realistic conditions we find a spontaneous emergence of disparate charge densities on the two surfaces. The surface charge densities can differ not only in their magnitude, but quite unexpectedly, even…
▽ More
The interaction between two chemically identical charge-regulated surfaces is studied using the classical density functional theory. In contrast to common expectations and assumptions, under certain realistic conditions we find a spontaneous emergence of disparate charge densities on the two surfaces. The surface charge densities can differ not only in their magnitude, but quite unexpectedly, even in their sign, implying that the electrostatic interaction between the two chemically identical surfaces can be attractive instead of repulsive. Moreover, an initial symmetry with equal charge densities on both surfaces can also be broken spontaneously upon decreasing the separation between the two surfaces. The origin of this phenomenon is a competition between the adsorption of ions from the solution to the surface and the interaction between the adsorbed ions already on the surface.These findings are fundamental for the understanding of the forces between colloidal objects and, in particular, they are bound to strongly influence the present picture of protein interaction.
△ Less
Submitted 23 February, 2018; v1 submitted 14 September, 2017;
originally announced September 2017.
-
Convergence of a flux-splitting finite volume scheme for conservation laws driven by Lévy noise
Authors:
Ananta K. Majee
Abstract:
We explore numerical approximation of multidimensional stochastic balance laws driven by multiplicative Lévy noise via flux- splitting finite volume method. The convergence of the approximations is proved towards the unique entropy solution of the underlying problem.
We explore numerical approximation of multidimensional stochastic balance laws driven by multiplicative Lévy noise via flux- splitting finite volume method. The convergence of the approximations is proved towards the unique entropy solution of the underlying problem.
△ Less
Submitted 10 August, 2017;
originally announced August 2017.
-
Optimal Strong Rates of Convergence for a Space-Time Discretization of the Stochastic Allen-Cahn Equation with multiplicative noise
Authors:
Ananta K. Majee,
Andreas Prohl
Abstract:
The stochastic Allen-Cahn equation with multiplicative noise involves the nonlinear drift operator ${\mathscr A}(x) = Δx - \bigl(\vert x\vert^2 -1\bigr)x$. We use the fact that ${\mathscr A}(x) = -{\mathcal J}^{\prime}(x)$ satisfies a weak monotonicity property to deduce uniform bounds in strong norms for solutions of the temporal, as well as of the spatio-temporal discretization of the problem. T…
▽ More
The stochastic Allen-Cahn equation with multiplicative noise involves the nonlinear drift operator ${\mathscr A}(x) = Δx - \bigl(\vert x\vert^2 -1\bigr)x$. We use the fact that ${\mathscr A}(x) = -{\mathcal J}^{\prime}(x)$ satisfies a weak monotonicity property to deduce uniform bounds in strong norms for solutions of the temporal, as well as of the spatio-temporal discretization of the problem. This weak monotonicity property then allows for the estimate $ \underset{1 \leq j \leq J}\sup {\mathbb E}\bigl[ \Vert X_{t_j} - Y^j\Vert_{{\mathbb L}^2}^2\bigr] \leq C_δ(k^{1-δ} + h^2)$ for all small $δ>0$, where $X$ is the strong variational solution of the stochastic Allen-Cahn equation, while $\big\{Y^j:0\le j\le J\big\}$ solves a structure preserving finite element based space-time discretization of the problem on a temporal mesh $\{ t_j;\, 1 \leq j \leq J\}$ of size $k>0$ which covers $[0,T]$.
△ Less
Submitted 28 May, 2017;
originally announced May 2017.
-
Rotational motion of dimers of Janus particles
Authors:
Arghya Majee
Abstract:
We theoretically study the motion of a rigid dimer of self-propelling Janus particles. In a simple kinetic approach without hydrodynamic interactions, the dimer moves on a helical trajectory and, at the same time, it rotates about its center of mass. Inclusion of the effects of mutual advection using superposition approximation does not alter the qualitative features of the motion but merely chang…
▽ More
We theoretically study the motion of a rigid dimer of self-propelling Janus particles. In a simple kinetic approach without hydrodynamic interactions, the dimer moves on a helical trajectory and, at the same time, it rotates about its center of mass. Inclusion of the effects of mutual advection using superposition approximation does not alter the qualitative features of the motion but merely changes the parameters of the trajectory and the angular velocity.
△ Less
Submitted 27 March, 2017;
originally announced March 2017.
-
Nanoscale Seebeck effect at hot metal nanostructures
Authors:
Aboubakry Ly,
Arghya Majee,
Alois Würger
Abstract:
We theoretically study the electrolyte Seebeck effect in the vicinity of a heated metal nanostructure, such as the cap of an active Janus colloid in an electrolyte, or gold- coated interfaces in optofluidic devices. The thermocharge accumulated at the surface varies with the local temperature, thus modulating the diffuse part of the electric double layer. On a conducting surface with non-uniform t…
▽ More
We theoretically study the electrolyte Seebeck effect in the vicinity of a heated metal nanostructure, such as the cap of an active Janus colloid in an electrolyte, or gold- coated interfaces in optofluidic devices. The thermocharge accumulated at the surface varies with the local temperature, thus modulating the diffuse part of the electric double layer. On a conducting surface with non-uniform temperature, the isopotential condition imposes a significant polarization charge within the metal. Surprisingly, this does not affect the slip velocity, which takes the same value on insulating and conducting surfaces. Our results for specific-ion effects agree qualitatively with recent observations for Janus colloids in different electrolyte solutions. Comparing the thermal, hydrodynamic, and ion diffusion time scales, we expect a rich transient behavior at the onset of thermally powered swimming, extending to microseconds after switching on the heating.
△ Less
Submitted 26 February, 2018; v1 submitted 9 March, 2017;
originally announced March 2017.
-
Poisson-Boltzmann study of the effective electrostatic interaction between colloids at an electrolyte interface
Authors:
Arghya Majee,
Markus Bier,
S. Dietrich
Abstract:
The effective electrostatic interaction between a pair of colloids, both of them located close to each other at an electrolyte interface, is studied by employing the full, nonlinear Poisson-Boltzmann (PB) theory within classical density functional theory. Using a simplified yet appropriate model, all contributions to the effective interaction are obtained exactly, albeit numerically. The compariso…
▽ More
The effective electrostatic interaction between a pair of colloids, both of them located close to each other at an electrolyte interface, is studied by employing the full, nonlinear Poisson-Boltzmann (PB) theory within classical density functional theory. Using a simplified yet appropriate model, all contributions to the effective interaction are obtained exactly, albeit numerically. The comparison between our results and those obtained within linearized PB theory reveals that the latter overestimates these contributions significantly at short inter-particle separations. Whereas the surface contributions to the linear and the nonlinear PB results differ only quantitatively, the line contributions show qualitative differences at short separations. Moreover, a dependence of the line contribution on the solvation properties of the two adjacent fluids is found, which is absent within the linear theory. Our results are expected to enrich the understanding of effective interfacial interactions between colloids.
△ Less
Submitted 2 August, 2016; v1 submitted 13 June, 2016;
originally announced June 2016.
-
A finite difference scheme for conservation laws driven by Levy noise
Authors:
Ujjwal Koley,
Ananta K. Majee,
Guy Vallet
Abstract:
In this paper, we analyze a semi-discrete finite difference scheme for a conservation laws driven by a homogeneous multiplicative Levy noise. Thanks to BV estimates, we show a compact sequence of approximate solutions, generated by the finite difference scheme, converges to the unique entropy solution of the underlying problem, as the spatial mesh size \Dx-->0. Moreover, we show that the expected…
▽ More
In this paper, we analyze a semi-discrete finite difference scheme for a conservation laws driven by a homogeneous multiplicative Levy noise. Thanks to BV estimates, we show a compact sequence of approximate solutions, generated by the finite difference scheme, converges to the unique entropy solution of the underlying problem, as the spatial mesh size \Dx-->0. Moreover, we show that the expected value of the L^1-difference between the approximate solution and the unique entropy solution converges at a rate O(\sqrt{\Dx}).
△ Less
Submitted 26 April, 2016;
originally announced April 2016.
-
Continuous dependence estimate for a degenerate parabolic-hyperbolic equation with Levy noise
Authors:
Ujjwal Koley,
Ananta K. Majee,
Guy Vallet
Abstract:
In this article, we are concerned with a multidimensional degenerate parabolic-hyperbolic equation driven by Levy processes. Using bounded variation (BV) estimates for vanishing viscosity approximations, we derive an explicit continuous dependence estimate on the nonlinearities of the entropy solutions under the assumption that Levy noise depends only on the solution. This result is used to show t…
▽ More
In this article, we are concerned with a multidimensional degenerate parabolic-hyperbolic equation driven by Levy processes. Using bounded variation (BV) estimates for vanishing viscosity approximations, we derive an explicit continuous dependence estimate on the nonlinearities of the entropy solutions under the assumption that Levy noise depends only on the solution. This result is used to show the error estimate for the stochastic vanishing viscosity method. In addition, we establish fractional BV estimate for vanishing viscosity approximations in case the noise coefficients depend on both the solution and spatial variable.
△ Less
Submitted 26 April, 2016;
originally announced April 2016.
-
On the Cauchy problem of a degenerate parabolic-hyperbolic PDE with Lévy noise
Authors:
Imran H. Biswas,
Ananta K. Majee,
Guy Vallet
Abstract:
In this article we deal with stochastic perturbation of degenerate parabolic partial differential equations (PDEs). The particular emphasise is on analysing the effect of multiplicative Lévy noise to such problems and establishing wellposedness by develo** a suitable weak entropy solution framework. The proof of existence is based on the vanishing viscosity technique. The uniqueness is settled b…
▽ More
In this article we deal with stochastic perturbation of degenerate parabolic partial differential equations (PDEs). The particular emphasise is on analysing the effect of multiplicative Lévy noise to such problems and establishing wellposedness by develo** a suitable weak entropy solution framework. The proof of existence is based on the vanishing viscosity technique. The uniqueness is settled by interpreting Kruzkov's doubling technique in the presence noise.
△ Less
Submitted 18 April, 2016;
originally announced April 2016.
-
Continuous dependence estimate for conservation laws with Lévy noise
Authors:
Imran H. Biswas,
Ujjwal Koley,
Ananta K. Majee
Abstract:
We are concerned with multidimensional stochastic balance laws driven by Lévy processes. Using bounded variation (BV) estimates for vanishing viscosity approximations, we derive an explicit continuous dependence estimate on the nonlinearities of the entropy solutions under the assumption that Lévy noise only depends on the solution. This result is used to show the error estimate for the stochastic…
▽ More
We are concerned with multidimensional stochastic balance laws driven by Lévy processes. Using bounded variation (BV) estimates for vanishing viscosity approximations, we derive an explicit continuous dependence estimate on the nonlinearities of the entropy solutions under the assumption that Lévy noise only depends on the solution. This result is used to show the error estimate for the stochastic vanishing viscosity method. In addition, we establish fractional $BV$ estimate for vanishing viscosity approximations in case the noise coefficient depends on both the solution and spatial variable.
△ Less
Submitted 9 February, 2015;
originally announced February 2015.
-
Conservation laws driven by Lévy white noise
Authors:
Imran H. Biswas,
Kenneth H. Karlsen,
Ananta K. Majee
Abstract:
We consider multidimensional conservation laws perturbed by multiplicative Lévy noise. We establish existence and uniqueness results for entropy solutions. The entropy inequalities are formally obtained by the Itô-Lévy chain rule. The multidimensionality requires a generalized interpretation of the entropy inequalities to accommodate Young measure-valued solutions. We first establish the existence…
▽ More
We consider multidimensional conservation laws perturbed by multiplicative Lévy noise. We establish existence and uniqueness results for entropy solutions. The entropy inequalities are formally obtained by the Itô-Lévy chain rule. The multidimensionality requires a generalized interpretation of the entropy inequalities to accommodate Young measure-valued solutions. We first establish the existence of entropy solutions in the generalized sense via the vanishing viscosity method, and then establish the $L^1$-contraction principle. Finally, the $L^1$ contraction principle is used to argue that the generalized entropy solution is indeed the classical entropy solution.
△ Less
Submitted 7 June, 2014;
originally announced June 2014.
-
Electrostatic interaction between colloidal particles trapped at an electrolyte interface
Authors:
Arghya Majee,
Markus Bier,
S. Dietrich
Abstract:
The electrostatic interaction between colloidal particles trapped at the interface between two immiscible electrolyte solutions is studied in the limit of small inter-particle distances. Within an appropriate model exact analytic expressions for the electrostatic potential as well as for the surface and line interaction energies are obtained. They demonstrate that the widely used superposition app…
▽ More
The electrostatic interaction between colloidal particles trapped at the interface between two immiscible electrolyte solutions is studied in the limit of small inter-particle distances. Within an appropriate model exact analytic expressions for the electrostatic potential as well as for the surface and line interaction energies are obtained. They demonstrate that the widely used superposition approximation, which is commonly applied to large distances between the colloidal particles, fails qualitatively at small distances and is quantitatively unreliable even at large distances. Our results contribute to an improved description of the interaction between colloidal particles trapped at fluid interfaces.
△ Less
Submitted 27 February, 2014;
originally announced February 2014.
-
Collective thermoelectrophoresis of charged colloids
Authors:
Arghya Majee,
Alois Würger
Abstract:
Thermally driven colloidal transport is, to a large extent, due to the thermoelectric or Seebeck effect of the charged solution.We show that, contrary to the generally adopted single-particle picture, the transport coefficient depends on the colloidal concentration. For solutions that are dilute in the hydrodynamic sense, collective effects may significantly affect the thermophoretic mobility. Our…
▽ More
Thermally driven colloidal transport is, to a large extent, due to the thermoelectric or Seebeck effect of the charged solution.We show that, contrary to the generally adopted single-particle picture, the transport coefficient depends on the colloidal concentration. For solutions that are dilute in the hydrodynamic sense, collective effects may significantly affect the thermophoretic mobility. Our results provide an explanation for recent experimental observations on polyelectrolytes and charged particles and suggest that for charged colloids collective behavior is the rule rather than the exception.
△ Less
Submitted 29 January, 2014;
originally announced January 2014.
-
Thermocharge of a hot spot in an electrolyte solution
Authors:
Arghya Majee,
Alois Wüurger
Abstract:
We discuss the thermoelectric properties of a locally heated micron-size volume in an electrolyte solution. We find that such a hot spot carries a net charge Q which, for an excess temperature of 10 K, may attain hundreds of elementary charges. The corresponding Seebeck electric field E increases linearly with the radius r inside the heated area, then goes through a maximum, and decays as 1/r2 at…
▽ More
We discuss the thermoelectric properties of a locally heated micron-size volume in an electrolyte solution. We find that such a hot spot carries a net charge Q which, for an excess temperature of 10 K, may attain hundreds of elementary charges. The corresponding Seebeck electric field E increases linearly with the radius r inside the heated area, then goes through a maximum, and decays as 1/r2 at larger distances. Our results could be relevant to optothermal actuation of electrolytes and colloidal suspensions.
△ Less
Submitted 29 January, 2014;
originally announced January 2014.
-
Charging of Heated Colloidal Particles Using the Electrolyte Seebeck Effect
Authors:
Arghya Majee,
Alois Würger
Abstract:
We propose a novel actuation mechanism for colloids, which is based on the Seebeck effect of the electrolyte solution: Laser heating of a nonionic particle accumulates in its vicinity a net charge Q, which is proportional to the excess temperature at the particle surface. The corresponding long-range thermoelectric field provides a tool for controlled interactions with nearby beads or with additio…
▽ More
We propose a novel actuation mechanism for colloids, which is based on the Seebeck effect of the electrolyte solution: Laser heating of a nonionic particle accumulates in its vicinity a net charge Q, which is proportional to the excess temperature at the particle surface. The corresponding long-range thermoelectric field provides a tool for controlled interactions with nearby beads or with additional molecular solutes. An external field Eext drags the thermocharged particle at a velocity that depends on its size and absorption properties; the latter point could be particularly relevant for separating carbon nanotubes according to their electronic band structure.
△ Less
Submitted 29 January, 2014;
originally announced January 2014.
-
Flow pattern in the vicinity of self-propelling hot Janus particles
Authors:
Thomas Bickel,
Arghya Majee,
Alois Würger
Abstract:
We study the temperature field and the resulting flow pattern in the vicinity of a heated metal-capped Janus particle. If its thickness exceeds about 10 nm, the cap forms an isotherm and the flow pattern comprises a quadrupolar term that decays with the square of the inverse distance 1/r^2. For much thinner caps the velocity varies as 1/r^3. These findings could be relevant for collective effects…
▽ More
We study the temperature field and the resulting flow pattern in the vicinity of a heated metal-capped Janus particle. If its thickness exceeds about 10 nm, the cap forms an isotherm and the flow pattern comprises a quadrupolar term that decays with the square of the inverse distance 1/r^2. For much thinner caps the velocity varies as 1/r^3. These findings could be relevant for collective effects in dense suspensions and for the circular tracer motion observed recently in the vicinity of a tethered Janus particle.
△ Less
Submitted 28 January, 2014;
originally announced January 2014.