arXiv subjects

Saurav Prakash

Publications and source records attributed to Saurav Prakash.

35 records · Page 2Linked to original sources

arXiv2023

Lottery Aware Sparsity Hunting: Enabling Federated Learning on Resource-Limited Edge

Edge devices can benefit remarkably from federated learning due to their distributed nature; however, their limited resource and computing power poses limitations in deployment. A possible solution to this problem is to utilize off-the-shelf sparse learning algorithms at the clients to meet their resource budget. However, such naive deployment in the clients causes significant accuracy degradation, especially for highly resource-constrained clients. In particular, our investigations reveal that the lack of consensus in the sparsity masks among the clients may potentially slow down the convergence of the global model and cause a substantial accuracy drop. With these observations, we present \textit{federated lottery aware sparsity hunting} (FLASH), a unified sparse learning framework for training a sparse sub-model that maintains the performance under ultra-low parameter density while yielding proportional communication benefits. Moreover, given that different clients may have different resource budgets, we present \textit{hetero-FLASH} where clients can take different density budgets based on their device resource limitations instead of supporting only one target parameter density. Experimental analysis on diverse models and datasets shows the superiority of FLASH in closing the gap with an unpruned baseline while yielding up to $\mathord{\sim}10.1\%$ improved accuracy with $\mathord{\sim}10.26\times$ fewer communication, compared to existing alternatives, at similar hyperparameter settings. Code is available at \url{https://github.com/SaraBabakN/flash_fl}.

cs.LG↗

arXiv2023

Federated Learning of Large Models at the Edge via Principal Sub-Model Training

Federated Learning (FL) is emerging as a popular, promising decentralized learning framework that enables collaborative training among clients, with no need to share private data between them or to a centralized server. However, considering many edge clients do not have sufficient computing, memory, or communication capabilities, federated learning of large models still faces significant bottlenecks. To keep such weak but crucial clients in the loop, prior works either consider a heterogeneous-client setting where clients train models with different sizes; or offload training to the server. However, the heterogeneous-client setting requires some clients to train full model, which is not aligned with the resource-constrained setting; while the latter ones break privacy promises in FL when sharing intermediate representations or labels with the server. To overcome these limitations, in this work, we formulate a realistic, but much less explored, cross-device FL setting in which no client can train a full large model nor is willing to share any intermediate information with the remote server. Under such a formulation, we develop a principal sub-model (PriSM) training methodology to collaboratively train a full large model, while assigning each client a small sub-model that is a probabilistic low-rank approximation to the full server model. When creating sub-models, PriSM first performs a principal kernel analysis in the orthogonal kernel space to obtain importance of each kernel. Then, PriSM adopts a novel importance-aware sampling process to select a subset of kernels (i.e., a kernel with high importance is assigned with a higher sampling probability). This sampling process ensures each sub-model is still a low-rank approximation to the full model, while all sub-models together achieve nearly full coverage on the principal kernels.

cs.LG↗

arXiv2023

Machine Unlearning of Federated Clusters

Federated clustering (FC) is an unsupervised learning problem that arises in a number of practical applications, including personalized recommender and healthcare systems. With the adoption of recent laws ensuring the "right to be forgotten", the problem of machine unlearning for FC methods has become of significant importance. We introduce, for the first time, the problem of machine unlearning for FC, and propose an efficient unlearning mechanism for a customized secure FC framework. Our FC framework utilizes special initialization procedures that we show are well-suited for unlearning. To protect client data privacy, we develop the secure compressed multiset aggregation (SCMA) framework that addresses sparse secure federated learning (FL) problems encountered during clustering as well as more general problems. To simultaneously facilitate low communication complexity and secret sharing protocols, we integrate Reed-Solomon encoding with special evaluation points into our SCMA pipeline, and prove that the client communication cost is logarithmic in the vector dimension. Additionally, to demonstrate the benefits of our unlearning mechanism over complete retraining, we provide a theoretical analysis for the unlearning performance of our approach. Simulation results show that the new FC framework exhibits superior clustering performance compared to previously reported FC baselines when the cluster sizes are highly imbalanced. Compared to completely retraining K-means++ locally and globally for each removal request, our unlearning procedure offers an average speed-up of roughly 84x across seven datasets. Our implementation for the proposed method is available at https://github.com/thupchnsky/mufc.

cs.LG↗

arXiv2023

Spatially reconfigurable topological textures in freestanding antiferromagnetic nanomembranes

Antiferromagnets hosting real-space topological spin textures are promising platforms to model fundamental ultrafast phenomena and explore spintronics. However, to date, they have only been fabricated epitaxially on specific symmetry-matched crystalline substrates, to preserve their intrinsic magneto-crystalline order. This curtails their integration with dissimilar supports, markedly restricting the scope of fundamental and applied investigations. Here, we circumvent this limitation by designing detachable crystalline antiferromagnetic nanomembranes of $α$-Fe$_{2}$O$_{3}$, that can be transferred onto other desirable supports after growth. We develop transmission-based antiferromagnetic vector-mapping to show that these nanomembranes harbour rich topological phenomenology at room temperature. Moreover, we exploit their extreme flexibility to demonstrate three-dimensional reconfiguration of antiferromagnetic properties, driven locally via flexure-induced strains. This allows us to spatially design antiferromagnetic states outside their typical thermal stability window. Integration of such freestanding antiferromagnetic layers with flat or curved nanostructures could enable spin texture designs tailored by magnetoelastic-/geometric-effects in the quasi-static and dynamical regimes, opening new explorations into curvilinear antiferromagnetism and unconventional computing.

cond-mat.mtrl-sci↗

arXiv2022

Experimental Evidence of t2g Electron-Gas Rashba Interaction Induced by Asymmetric Orbital Hybridization

We report the control of Rashba spin-orbit interaction by tuning asymmetric hybridization between Ti-orbitals at the LaAlO3/SrTiO3 interface. This asymmetric orbital hybridization is modulated by introducing a LaFeO3 layer between LaAlO3 and SrTiO3, which alters the Ti-O lattice polarization and traps interfacial charge carriers, resulting in a large Rashba spin-orbit effect at the interface in the absence of an external bias. This observation is verified through high-resolution electron microscopy, magneto-transport and first-principles calculations. Our results open hitherto unexplored avenues of controlling Rashba interaction to design next-generation spin-orbitronics.

cond-mat.mtrl-sci↗

arXiv2022

Basil: A Fast and Byzantine-Resilient Approach for Decentralized Training

Detection and mitigation of Byzantine behaviors in a decentralized learning setting is a daunting task, especially when the data distribution at the users is heterogeneous. As our main contribution, we propose Basil, a fast and computationally efficient Byzantine robust algorithm for decentralized training systems, which leverages a novel sequential, memory assisted and performance-based criteria for training over a logical ring while filtering the Byzantine users. In the IID dataset distribution setting, we provide the theoretical convergence guarantees of Basil, demonstrating its linear convergence rate. Furthermore, for the IID setting, we experimentally demonstrate that Basil is robust to various Byzantine attacks, including the strong Hidden attack, while providing up to ${\sim}16 \%$ higher test accuracy over the state-of-the-art Byzantine-resilient decentralized learning approach. Additionally, we generalize Basil to the non-IID dataset distribution setting by proposing Anonymous Cyclic Data Sharing (ACDS), a technique that allows each node to anonymously share a random fraction of its local non-sensitive dataset (e.g., landmarks images) with all other nodes. We demonstrate that Basil alongside ACDS with only $5\%$ data sharing provides effective toleration of Byzantine nodes, unlike the state-of-the-art Byzantine robust algorithm that completely fails in the heterogeneous data setting. Finally, to reduce the overall latency of Basil resulting from its sequential implementation over the logical ring, we propose Basil+. In particular, Basil+ provides scalability by enabling Byzantine-robust parallel training across groups of logical rings, and at the same time, it retains the performance gains of Basil due to sequential training within each group. Furthermore, we experimentally demonstrate the scalability gains of Basil+ through different sets of experiments.

eess.SY↗

arXiv2022

Secure and Fault Tolerant Decentralized Learning

Federated learning (FL) is a promising paradigm for training a global model over data distributed across multiple data owners without centralizing clients' raw data. However, sharing of local model updates can also reveal information of clients' local datasets. Trusted execution environments (TEEs) within the FL server have been recently deployed by companies like Meta for secure aggregation. However, secure aggregation can suffer from error-prone local updates sent by clients that become faulty during training due to underlying device malfunctions. Also, data heterogeneity across clients makes fault mitigation challenging, as even updates from normal clients are dissimilar. Thus, most of the prior fault tolerant methods, which treat any local update differing from the majority of other updates as faulty, perform poorly. We propose DiverseFL to make model aggregation secure as well as robust to faults. In DiverseFL, any client whose local model update diverges from its associated guiding update is tagged as being faulty. To implement our novel per-client criteria for fault mitigation, DiverseFL creates a TEE-based secure enclave within the FL server, which in addition to performing secure aggregation for carrying out the global model update step, securely receives a small representative sample of local data from each client only once before training, and computes guiding updates for each participating client during training. Thus, DiverseFL provides security against privacy leakage as well as robustness against faulty clients. In experiments, DiverseFL consistently achieves significant improvements in absolute test accuracy over prior fault mitigation benchmarks. DiverseFL also performs closely to OracleSGD, where server combines updates only from the normal clients. We also analyze the convergence rate of DiverseFL under non-IID data and standard convexity assumptions.

cs.DC↗

arXiv2021

CodedReduce: A Fast and Robust Framework for Gradient Aggregation in Distributed Learning

We focus on the commonly used synchronous Gradient Descent paradigm for large-scale distributed learning, for which there has been a growing interest to develop efficient and robust gradient aggregation strategies that overcome two key system bottlenecks: communication bandwidth and stragglers' delays. In particular, Ring-AllReduce (RAR) design has been proposed to avoid bandwidth bottleneck at any particular node by allowing each worker to only communicate with its neighbors that are arranged in a logical ring. On the other hand, Gradient Coding (GC) has been recently proposed to mitigate stragglers in a master-worker topology by allowing carefully designed redundant allocation of the data set to the workers. We propose a joint communication topology design and data set allocation strategy, named CodedReduce (CR), that combines the best of both RAR and GC. That is, it parallelizes the communications over a tree topology leading to efficient bandwidth utilization, and carefully designs a redundant data set allocation and coding strategy at the nodes to make the proposed gradient aggregation scheme robust to stragglers. In particular, we quantify the communication parallelization gain and resiliency of the proposed CR scheme, and prove its optimality when the communication topology is a regular tree. Moreover, we characterize the expected run-time of CR and show order-wise speedups compared to the benchmark schemes. Finally, we empirically evaluate the performance of our proposed CR design over Amazon EC2 and demonstrate that it achieves speedups of up to 27.2x and 7.0x, respectively over the benchmarks GC and RAR.

stat.ML↗

arXiv2021

Coded Computing for Federated Learning at the Edge

Federated Learning (FL) is an exciting new paradigm that enables training a global model from data generated locally at the client nodes, without moving client data to a centralized server. Performance of FL in a multi-access edge computing (MEC) network suffers from slow convergence due to heterogeneity and stochastic fluctuations in compute power and communication link qualities across clients. A recent work, Coded Federated Learning (CFL), proposes to mitigate stragglers and speed up training for linear regression tasks by assigning redundant computations at the MEC server. Coding redundancy in CFL is computed by exploiting statistical properties of compute and communication delays. We develop CodedFedL that addresses the difficult task of extending CFL to distributed non-linear regression and classification problems with multioutput labels. The key innovation of our work is to exploit distributed kernel embedding using random Fourier features that transforms the training task into distributed linear regression. We provide an analytical solution for load allocation, and demonstrate significant performance gains for CodedFedL through experiments over benchmark datasets using practical network parameters.

cs.DC↗

arXiv2021

Coded Computing for Low-Latency Federated Learning over Wireless Edge Networks

Federated learning enables training a global model from data located at the client nodes, without data sharing and moving client data to a centralized server. Performance of federated learning in a multi-access edge computing (MEC) network suffers from slow convergence due to heterogeneity and stochastic fluctuations in compute power and communication link qualities across clients. We propose a novel coded computing framework, CodedFedL, that injects structured coding redundancy into federated learning for mitigating stragglers and speeding up the training procedure. CodedFedL enables coded computing for non-linear federated learning by efficiently exploiting distributed kernel embedding via random Fourier features that transforms the training task into computationally favourable distributed linear regression. Furthermore, clients generate local parity datasets by coding over their local datasets, while the server combines them to obtain the global parity dataset. Gradient from the global parity dataset compensates for straggling gradients during training, and thereby speeds up convergence. For minimizing the epoch deadline time at the MEC server, we provide a tractable approach for finding the amount of coding redundancy and the number of local data points that a client processes during training, by exploiting the statistical properties of compute as well as communication delays. We also characterize the leakage in data privacy when clients share their local parity datasets with the server. We analyze the convergence rate and iteration complexity of CodedFedL under simplifying assumptions, by treating CodedFedL as a stochastic gradient descent algorithm. Furthermore, we conduct numerical experiments using practical network parameters and benchmark datasets, where CodedFedL speeds up the overall training time by up to $15\times$ in comparison to the benchmark schemes.

cs.DC↗

arXiv2020

Coded Federated Learning

Federated learning is a method of training a global model from decentralized data distributed across client devices. Here, model parameters are computed locally by each client device and exchanged with a central server, which aggregates the local models for a global view, without requiring sharing of training data. The convergence performance of federated learning is severely impacted in heterogeneous computing platforms such as those at the wireless edge, where straggling computations and communication links can significantly limit timely model parameter updates. This paper develops a novel coded computing technique for federated learning to mitigate the impact of stragglers. In the proposed Coded Federated Learning (CFL) scheme, each client device privately generates parity training data and shares it with the central server only once at the start of the training phase. The central server can then preemptively perform redundant gradient computations on the composite parity data to compensate for the erased or delayed parameter updates. Our results show that CFL allows the global model to converge nearly four times faster when compared to an uncoded approach

cs.LG↗

arXiv2020

Antiferromagnetic Half-skyrmions and Bimerons at room temperature

In the quest for post-CMOS technologies, ferromagnetic skyrmions and their anti-particles have shown great promise as topologically protected solitonic information carriers in memory-in-logic or neuromorphic devices. However, the presence of dipolar fields in ferromagnets, restricting the formation of ultra-small topological textures, and the deleterious skyrmion Hall effect when driven by spin torques have thus far inhibited their practical implementations. Antiferromagnetic analogues, which are predicted to demonstrate relativistic dynamics, fast deflection-free motion and size scaling have recently come into intense focus, but their experimental realizations in natural antiferromagnetic systems are yet to emerge. Here, we demonstrate a family of topological antiferromagnetic spin-textures in $α$-Fe$_2$O$_3$ - an earth-abundant oxide insulator - capped with a Pt over-layer. By exploiting a first-order analogue of the Kibble-Zurek mechanism, we stabilize exotic merons-antimerons (half-skyrmions), and bimerons, which can be erased by magnetic fields and re-generated by temperature cycling. These structures have characteristic sizes of the order ~100 nm that can be chemically controlled via precise tuning of the exchange and anisotropy, with pathways to further scaling. Driven by current-based spin torques from the heavy-metal over-layer, some of these AFM textures could emerge as prime candidates for low-energy antiferromagnetic spintronics at room temperature.

cond-mat.mtrl-sci↗

arXiv2020

Coded Computing for Distributed Graph Analytics

Performance of distributed graph processing systems significantly suffers from 'communication bottleneck' as a large number of messages are exchanged among servers at each step of the computation. Motivated by graph based MapReduce, we propose a coded computing framework that leverages computation redundancy to alleviate the communication bottleneck in distributed graph processing. We develop a novel 'coding' scheme that systematically injects structured redundancy in computation phase to enable 'coded' multicasting opportunities during message exchange between servers, reducing communication load substantially in large-scale graph processing. For theoretical analysis, we consider random graph models, and prove that our proposed scheme enables an (asymptotically) inverse-linear trade-off between 'computation load' and 'average communication load' for two popular random graph models -- Erdos-Renyi model, and power law model. Particularly, for a given computation load r, (i.e. when each graph vertex is carefully stored at r servers), the proposed scheme slashes the average communication load by (nearly) a multiplicative factor of r. For the Erdos-Renyi model, our proposed scheme is optimal asymptotically as the graph size increases by providing an information-theoretic converse. To illustrate the benefits of our scheme in practice, we implement PageRank over Amazon EC2, using artificial as well as real-world datasets, demonstrating significant gains over conventional PageRank. We also specialize our scheme and extend our theoretical results to two other random graph models -- random bi-partite model, and stochastic block model. They asymptotically enable inverse-linear trade-offs between computation and communication loads in distributed graph processing for these popular random graph models as well. We complement the achievability results with converse bounds for both of these models.

cs.DC↗

arXiv2020

Characteristic Lengths of Interlayer Charge-Transfer in Correlated Oxide Heterostructures

Using interlayer interaction to control functional heterostructures with atomic-scale designs has become one of the most effective interface-engineering strategies nowadays. Here, we demonstrate the effect of a crystalline LaFeO3 buffer layer on amorphous and crystalline LaAlO3/SrTiO3 heterostructures. The LaFeO3 buffer layer acts as an energetically favored electron acceptor in both LaAlO3/SrTiO3 systems, resulting in modulation of interfacial carrier density and hence metal-to-insulator transition. For amorphous and crystalline LaAlO3/SrTiO3 heterostructures, the metal-to-insulator transition is found when the LaFeO3 layer thickness crosses 3 and 6 unit cells, respectively. Such different critical LaFeO3 thicknesses are explained in terms of distinct characteristic lengths of the redox-reaction-mediated and polar-catastrophe-dominated charge transfer, controlled by the interfacial atomic contact and Thomas-Fermi screening effect, respectively. Our results not only shed light on the complex interlayer charge transfer across oxide heterostructures but also provides a new route to precisely tailor the charge-transfer process at a functional interface.

cond-mat.mtrl-sci↗

arXiv2019

A Pre-defined Sparse Kernel Based Convolution for Deep CNNs

The high demand for computational and storage resources severely impede the deployment of deep convolutional neural networks (CNNs) in limited-resource devices. Recent CNN architectures have proposed reduced complexity versions (e.g. SuffleNet and MobileNet) but at the cost of modest decreases inaccuracy. This paper proposes pSConv, a pre-defined sparse 2D kernel-based convolution, which promises significant improvements in the trade-off between complexity and accuracy for both CNN training and inference. To explore the potential of this approach, we have experimented with two widely accepted datasets, CIFAR-10 and Tiny ImageNet, in sparse variants of both the ResNet18 and VGG16 architectures. Our approach shows a parameter count reduction of up to 4.24x with modest degradation in classification accuracy relative to that of standard CNNs. Our approach outperforms a popular variant of ShuffleNet using a variant of ResNet18 with pSConv having 3x3 kernels with only four of nine elements not fixed at zero. In particular, the parameter count is reduced by 1.7x for CIFAR-10 and 2.29x for Tiny ImageNet with an increased accuracy of ~4%.

cs.CV↗

arXiv2019

Coded Computation over Heterogeneous Clusters

In large-scale distributed computing clusters, such as Amazon EC2, there are several types of "system noise" that can result in major degradation of performance: bottlenecks due to limited communication bandwidth, latency due to straggler nodes, etc. On the other hand, these systems enjoy abundance of redundancy - a vast number of computing nodes and large storage capacity. There have been recent results that demonstrate the impact of coding for efficient utilization of computation and storage redundancy to alleviate the effect of stragglers and communication bottlenecks in homogeneous clusters. In this paper, we focus on general heterogeneous distributed computing clusters consisting of a variety of computing machines with different capabilities. We propose a coding framework for speeding up distributed computing in heterogeneous clusters by trading redundancy for reducing the latency of computation. In particular, we propose Heterogeneous Coded Matrix Multiplication (HCMM) algorithm for performing distributed matrix multiplication over heterogeneous clusters that is provably asymptotically optimal for a broad class of processing time distributions. Moreover, we show that HCMM is unboundedly faster than any uncoded scheme. To demonstrate practicality of HCMM, we carry out experiments over Amazon EC2 clusters where HCMM is found to be up to $61\%$, $46\%$ and $36\%$ respectively faster than three benchmark load allocation schemes - Uniform Uncoded, Load-balanced Uncoded, and Uniform Coded. Additionally, we provide a generalization to the problem of optimal load allocation in heterogeneous settings, where we take into account the monetary costs associated with the clusters. We argue that HCMM is asymptotically optimal for budget-constrained scenarios as well, and we develop a heuristic algorithm for (HCMM) load allocation for budget-limited computation tasks.

cs.DC↗

arXiv2018

Phase change-photonic framework for terahertz wave control

The advancement in capabilities enabled by handheld devices and the prospective internet of things has led to an explosion of wireless data consumption in the last decade. The ever-increasing demand for high speed data transfer has projected the unprecedented importance of terahertz spectral range for next generation wireless communication systems. However, versatile interaction and active manipulation of terahertz waves required for a wide-range of terahertz technologies are either unavailable or unscalable. Here, we report on an integrated platform of chalcogenide phase change material, germanium antimony telluride (GST), with metamaterials for multidimensional manipulation of terahertz waves. The thermal, electrical and optical stimulation of GST provide multilevel non-volatility, sub-wavelength spatial selectivity and variable ultrafast switching of terahertz waves, respectively. These multifaceted response of phase change material can be used in tandem with metamaterial that enables strong field confinement along with on-demand spectral response and frequency selectivity. The unique and multifarious properties of GST combined with the versatility and scalability of metamaterial design and ease of fabrication could aid in the development of advanced devices operating in terahertz range, such as spatio-temporal light modulators, neuromorphic photonic devices, compressive imagers, and switchable holograms, vital for the next generation wireless communication systems.

physics.app-ph↗