Offering incentives (e.g., coupons at Amazon, discounts at Uber and video bonuses at Tiktok) to user is a common strategy used by online platforms to increase user engagement and platform revenue. Despite its proven effectiveness, these marketing incentives incur an inevitable cost and might result in a low ROI (Return on Investment) if not used properly. On the other hand, different users respond differently to these incentives, for instance, some users never buy certain products without coupons, while others do anyway. Thus, how to select the right amount of incentives (i.e. treatment) to each user under budget constraints is an important research problem with great practical implications. In this paper, we call such problem as a budget-constrained treatment selection (BTS) problem. The challenge is how to efficiently solve BTS problem on a Large-Scale dataset and achieve improved results over the existing techniques. We propose a novel tree-based treatment selection technique under budget constraints, called Large-Scale Budget-Constrained Causal Forest (LBCF) algorithm, which is also an efficient treatment selection algorithm suitable for modern distributed computing systems. A novel offline evaluation method is also proposed to overcome an intrinsic challenge in assessing solutions' performance for BTS problem in randomized control trials (RCT) data. We deploy our approach in a real-world scenario on a large-scale video platform, where the platform gives away bonuses in order to increase users' campaign engagement duration. The simulation analysis, offline and online experiments all show that our method outperforms various tree-based state-of-the-art baselines. The proposed approach is currently serving over hundreds of millions of users on the platform and achieves one of the most tremendous improvements over these months.
In this paper, we propose a novel architecture for a deep learning system, named k-degree layer-wise network, to realize efficient geo-distributed computing between Cloud and Internet of Things (IoT). The geo-distributed computing extends Cloud to the geographical verge of the network in the neighbor of IoT. The basic ideas of the proposal include a k-degree constraint and a layer-wise constraint. The k-degree constraint is defined such that the degree of each vertex on the h-th layer is exactly k(h) to extend the existing deep belief networks and control the communication cost. The layer-wise constraint is defined such that the layer-wise degrees are monotonically decreasing in positive direction to gradually reduce the dimension of data. We prove the k-degree layer-wise network is sparse, while a typical deep neural network is dense. In an evaluation on the M-distributed MNIST database, the proposal is superior to a state-of-the-art model in terms of communication cost and learning time with scalability.
My project tackles the question of whether Ray can be used to quickly train autonomous vehicles using a simulator (Carla), and whether a platform robust enough for further research purposes can be built around it. Ray is an open-source framework that enables distributed machine learning applications. Distributed computing is a technique which parallelizes computational tasks, such as training a model, among many machines. Ray abstracts away the complex coordination of these machines, making it rapidly scalable. Carla is a vehicle simulator that generates data used to train a model. The bulk of the project was writing the training logic that Ray would use to train my distributed model. Imitation learning is the best fit for autonomous vehicles. Imitation learning is an alternative to reinforcement learning and it works by trying to learn the optimal policy by imitating an expert (usually a human) given a set of demonstrations. A key deliverable for the project was showcasing my trained agent in a few benchmark tests, such as navigating a complex turn through traffic. Beyond that, the broader ambition was to develop a research platform where others could quickly train and run experiments on huge amounts of Carla vehicle data. Thus, my end product is not a single model, but a large-scale, open-source research platform (RayCarla) for autonomous vehicle researchers to utilize.
Evolutionary algorithms provide gradient-free optimisation which is beneficial for models that have difficulty in obtaining gradients; for instance, geoscientific landscape evolution models. However, such models are at times computationally expensive and even distributed swarm-based optimisation with parallel computing struggles. We can incorporate efficient strategies such as surrogate-assisted optimisation to address the challenges; however, implementing inter-process communication for surrogate-based model training is difficult. In this paper, we implement surrogate-based estimation of fitness evaluation in distributed swarm optimisation over a parallel computing architecture. We first test the framework on a set of benchmark optimisation problems and then apply it to a geoscientific model that features a landscape evolution model. Our results demonstrate very promising results for benchmark functions and the Badlands landscape evolution model. We obtain a reduction in computational time while retaining optimisation solution accuracy through the use of surrogates in a parallel computing environment. The major contribution of the paper is in the application of surrogate-based optimisation for geoscientific models which can in the future help in a better understanding of paleoclimate and geomorphology.
In the modern era of computing, machine learning tools have demonstrated their potential in vital sectors, such as healthcare and finance, to derive proper inferences. The sensitive and confidential nature of the data in such sectors raises genuine concerns for data privacy. This motivated the area of Privacy-preserving Machine Learning (PPML), where privacy of data is guaranteed. In this thesis, we design an efficient platform, MPCLeague, for PPML in the Secure Outsourced Computation (SOC) setting using Secure Multi-party Computation (MPC) techniques. MPC, the holy-grail problem of secure distributed computing, enables a set of n mutually distrusting parties to perform joint computation on their private inputs in a way that no coalition of t parties can learn more information than the output (privacy) or affect the true output of the computation (correctness). While MPC, in general, has been a subject of extensive research, the area of MPC with a small number of parties has drawn popularity of late mainly due to its application to real-time scenarios, efficiency and simplicity. This thesis focuses on designing efficient MPC frameworks for 2, 3 and 4 parties, with at most one corruption and supports ring structures. At the heart of this thesis are four frameworks - ASTRA, SWIFT, Tetrad, ABY2.0 - catered to different settings. The practicality of our framework is argued through improvements in the benchmarking of widely used ML algorithms -- Linear Regression, Logistic Regression, Neural Networks, and Support Vector Machines. We propose two variants for each of our frameworks, with one variant aiming to minimise the execution time while the other focuses on the monetary cost. The concrete efficiency gains of our frameworks coupled with the stronger security guarantee of robustness make our platform an ideal choice for a real-time deployment of PPML techniques.
Workflow scheduling is a long-studied problem in parallel and distributed computing (PDC), aiming to efficiently utilize compute resources to meet user's service requirements. Recently proposed scheduling methods leverage the low response times of edge computing platforms to optimize application Quality of Service (QoS). However, scheduling workflow applications in mobile edge-cloud systems is challenging due to computational heterogeneity, changing latencies of mobile devices and the volatile nature of workload resource requirements. To overcome these difficulties, it is essential, but at the same time challenging, to develop a long-sighted optimization scheme that efficiently models the QoS objectives. In this work, we propose MCDS: Monte Carlo Learning using Deep Surrogate Models to efficiently schedule workflow applications in mobile edge-cloud computing systems. MCDS is an Artificial Intelligence (AI) based scheduling approach that uses a tree-based search strategy and a deep neural network-based surrogate model to estimate the long-term QoS impact of immediate actions for robust optimization of scheduling decisions. Experiments on physical and simulated edge-cloud testbeds show that MCDS can improve over the state-of-the-art methods in terms of energy consumption, response time, SLA violations and cost by at least 6.13, 4.56, 45.09 and 30.71 percent respectively.
One underlying assumption of recent federated learning (FL) paradigms is that all local models usually share the same network architecture and size, which becomes impractical for devices with different hardware resources. A scalable federated learning framework should address the heterogeneity that clients have different computing capacities and communication capabilities. To this end, this paper proposes FedHM, a novel heterogeneous federated model compression framework, distributing the heterogeneous low-rank models to clients and then aggregating them into a full-rank model. Our solution enables the training of heterogeneous models with varying computational complexities and aggregates them into a single global model. Furthermore, FedHM significantly reduces the communication cost by using low-rank models. Extensive experimental results demonstrate that FedHM is superior in the performance and robustness of models of different sizes, compared with state-of-the-art heterogeneous FL methods under various FL settings. Additionally, the convergence guarantee of FL for heterogeneous devices is first theoretically analyzed.
Realization of universal computing units for nanorobots is highly promising in creating new and wide arrays of applications, particularly in the realm of distributed computation. However, such realization is also a challenging problem due to the physical limitations of nanometer-sized designs such as in computation, sensory and perception as well as actuation. This paper proposes a theoretical foundation for solving this problem based on a novel notion of distributed swarm computing by basis agents (BAs). The proposed BA is an abstract model for nanorobots that can compute a very simple basis function called B-function. It is mathematically shown here that a swarm of BAs has the universal function approximation property and can accurately approximate functions. It is then analytically demonstrated that a swarm of BAs can be easily reprogrammed to compute desired functions simply by adjusting the concentrations of BAs in the environment. We further propose a specific structure for BAs which enable them to perform distributed computing such as in the aqueous environment of living tissues and nanomedicine. The hardware complexity of this structure aims to remain low to be more reasonably realizable by today technology. Finally, the performance of the proposed approach is illustrated by a simulation example.
Deep neural networks (DNNs) exploit many layers and a large number of parameters to achieve excellent performance. The training process of DNN models generally handles large-scale input data with many sparse features, which incurs high Input/Output (IO) cost, while some layers are compute-intensive. The training process generally exploits distributed computing resources to reduce training time. In addition, heterogeneous computing resources, e.g., CPUs, GPUs of multiple types, are available for the distributed training process. Thus, the scheduling of multiple layers to diverse computing resources is critical for the training process. To efficiently train a DNN model using the heterogeneous computing resources, we propose a distributed framework, i.e., Paddle-Heterogeneous Parameter Server (Paddle-HeterPS), composed of a distributed architecture and a Reinforcement Learning (RL)-based scheduling method. The advantages of Paddle-HeterPS are three-fold compared with existing frameworks. First, Paddle-HeterPS enables efficient training process of diverse workloads with heterogeneous computing resources. Second, Paddle-HeterPS exploits an RL-based method to efficiently schedule the workload of each layer to appropriate computing resources to minimize the cost while satisfying throughput constraints. Third, Paddle-HeterPS manages data storage and data communication among distributed computing resources. We carry out extensive experiments to show that Paddle-HeterPS significantly outperforms state-of-the-art approaches in terms of throughput (14.5 times higher) and monetary cost (312.3% smaller). The codes of the framework are publicly available at: https://github.com/PaddlePaddle/Paddle.
Echosounders are high-frequency sonar systems used to sense fish and zooplankton underwater. Their deployment on a variety of ocean observing platforms is generating vast amounts of data at an unprecedented speed from the oceans. Efficient and integrative analysis of these data, whether across different echosounder instruments or in combination with other oceanographic datasets, is crucial for understanding marine ecosystem response to the rapidly changing climate. Here we present Echopype, an open-source Python software library designed to address this need. By standardizing data as labeled, multi-dimensional arrays encoded in the widely embraced netCDF data model following a community convention, Echopype enhances the interoperability of echosounder data, making it easier to explore and use. By leveraging scientific Python libraries optimized for distributed computing, Echopype achieves computational scalability, enabling efficient processing in both local and cloud computing environments. Echopype's modularized package structure further provides a unified framework for expanding support for additional instrument raw data formats and incorporating new analysis functionalities. We plan to continue developing Echopype by supporting and collaborating with the echosounder user community, and envision that the growth of this package will catalyze the integration of echosounder data into broader regional and global ocean observation strategies.
We consider the classical problem of scheduling task graphs corresponding to complex applications on distributed computing systems. A number of heuristics have been previously proposed to optimize task scheduling with respect to metrics such as makespan and throughput. However, they tend to be slow to run, particularly for larger problem instances, limiting their applicability in more dynamic systems. Motivated by the goal of solving these problems more rapidly, we propose, for the first time, a graph convolutional network-based scheduler (GCNScheduler). By carefully integrating an inter-task data dependency structure with network settings into an input graph and feeding it to an appropriate GCN, the GCNScheduler can efficiently schedule tasks of complex applications for a given objective. We evaluate our scheme with baselines through simulations. We show that not only can our scheme quickly and efficiently learn from existing scheduling schemes, but also it can easily be applied to large-scale settings where current scheduling schemes fail to handle. We show that it achieves better makespan than the classic HEFT algorithm, and almost the same throughput as throughput-oriented HEFT (TP-HEFT), while providing several orders of magnitude faster scheduling times in both cases. For example, for makespan minimization, GCNScheduler schedules 50-node task graphs in about 4 milliseconds while HEFT takes more than 1500 seconds; and for throughput maximization, GCNScheduler schedules 100-node task graphs in about 3.3 milliseconds, compared to about 6.9 seconds for TP-HEFT.
Variational inequalities in general and saddle point problems in particular are increasingly relevant in machine learning applications, including adversarial learning, GANs, transport and robust optimization. With increasing data and problem sizes necessary to train high performing models across various applications, we need to rely on parallel and distributed computing. However, in distributed training, communication among the compute nodes is a key bottleneck during training, and this problem is exacerbated for high dimensional and over-parameterized models. Due to these considerations, it is important to equip existing methods with strategies that would allow to reduce the volume of transmitted information during training while obtaining a model of comparable quality. In this paper, we present the first theoretically grounded distributed methods for solving variational inequalities and saddle point problems using compressed communication: MASHA1 and MASHA2. Our theory and methods allow for the use of both unbiased (such as Rand$k$; MASHA1) and contractive (such as Top$k$; MASHA2) compressors. New algorithms support bidirectional compressions, and also can be modified for stochastic setting with batches and for federated learning with partial participation of clients. We empirically validated our conclusions using two experimental setups: a standard bilinear min-max problem, and large-scale distributed adversarial training of transformers.
Classic and deep generalized canonical correlation analysis (GCCA) algorithms seek low-dimensional common representations of data entities from multiple ``views'' (e.g., audio and image) using linear transformations and neural networks, respectively. When the views are acquired and stored at different computing agents (e.g., organizations and edge devices) and data sharing is undesired due to privacy or communication cost considerations, federated learning-based GCCA is well-motivated. In federated learning, the views are kept locally at the agents and only derived, limited information exchange with a central server is allowed. However, applying existing GCCA algorithms onto such federated learning settings may incur prohibitively high communication overhead. This work puts forth a communication-efficient federated learning framework for both linear and deep GCCA under the maximum variance (MAX-VAR) formulation. The overhead issue is addressed by aggressively compressing (via quantization) the exchanging information between the computing agents and a central controller. Compared to the unquantized version, our empirical study shows that the proposed algorithm enjoys a substantial reduction of communication overheads with virtually no loss in accuracy and convergence speed. Rigorous convergence analyses are also presented, which is a nontrivial effort. Generic federated optimization results do not cover the special problem structure of GCCA. Our result shows that the proposed algorithms for both linear and deep GCCA converge to critical points at a sublinear rate, even under heavy quantization and stochastic approximations. In addition, in the linear MAX-VAR case, the quantized algorithm approaches a global optimum in a geometric rate under reasonable conditions. Synthetic and real-data experiments are used to showcase the effectiveness of the proposed approach.
Distributed computing enables large-scale computation tasks to be processed over multiple workers in parallel. However, the randomness of communication and computation delays across workers causes the straggler effect, which may degrade the performance. Coded computation helps to mitigate the straggler effect, but the amount of redundant load and their assignment to the workers should be carefully optimized. In this work, we consider a multi-master heterogeneous-worker distributed computing scenario, where multiple matrix multiplication tasks are encoded and allocated to workers for parallel computation. The goal is to minimize the communication plus computation delay of the slowest task. We propose worker assignment, resource allocation and load allocation algorithms under both dedicated and fractional worker assignment policies, where each worker can process the encoded tasks of either a single master or multiple masters, respectively. Then, the non-convex delay minimization problem is solved by employing the Markov's inequality-based approximation, Karush-Kuhn-Tucker conditions, and successive convex approximation methods. Through extensive simulations, we show that the proposed algorithms can reduce the task completion delay compared to the benchmarks, and observe that dedicated and fractional worker assignment policies have different scopes of applications.
Data augmentation (DA) algorithms are widely used for Bayesian inference due to their simplicity. In massive data settings, however, DA algorithms are prohibitively slow because they pass through the full data in any iteration, imposing serious restrictions on their usage despite the advantages. Addressing this problem, we develop a framework for extending any DA that exploits asynchronous and distributed computing. The extended DA algorithm is indexed by a parameter $r \in (0, 1)$ and is called Asynchronous and Distributed (AD) DA with the original DA as its parent. Any ADDA starts by dividing the full data into $k$ smaller disjoint subsets and storing them on $k$ processes, which could be machines or processors. Every iteration of ADDA augments only an $r$-fraction of the $k$ data subsets with some positive probability and leaves the remaining $(1-r)$-fraction of the augmented data unchanged. The parameter draws are obtained using the $r$-fraction of new and $(1-r)$-fraction of old augmented data. For many choices of $k$ and $r$, the fractional updates of ADDA lead to a significant speed-up over the parent DA in massive data settings, and it reduces to the distributed version of its parent DA when $r=1$. We show that the ADDA Markov chain is Harris ergodic with the desired stationary distribution under mild conditions on the parent DA algorithm. We demonstrate the numerical advantages of the ADDA in three representative examples corresponding to different kinds of massive data settings encountered in applications. In all these examples, our DA generalization is significantly faster than its parent DA algorithm for all the choices of $k$ and $r$. We also establish geometric ergodicity of the ADDA Markov chain for all three examples, which in turn yields asymptotically valid standard errors for estimates of desired posterior quantities.
Adaptive gradient methods including Adam, AdaGrad, and their variants have been very successful for training deep learning models, such as neural networks. Meanwhile, given the need for distributed computing, distributed optimization algorithms are rapidly becoming a focal point. With the growth of computing power and the need for using machine learning models on mobile devices, the communication cost of distributed training algorithms needs careful consideration. In this paper, we introduce novel convergent decentralized adaptive gradient methods and rigorously incorporate adaptive gradient methods into decentralized training procedures. Specifically, we propose a general algorithmic framework that can convert existing adaptive gradient methods to their decentralized counterparts. In addition, we thoroughly analyze the convergence behavior of the proposed algorithmic framework and show that if a given adaptive gradient method converges, under some specific conditions, then its decentralized counterpart is also convergent. We illustrate the benefit of our generic decentralized framework on a prototype method, i.e., AMSGrad, both theoretically and numerically.
We consider the problem of simultaneous scheduling and resource allocation of an incoming flow of requests to a set of computing units. By representing each computing unit as a node, we model the overall system as a multi-queue scheme. Inspired by congestion control approaches in communication networks, we propose an AIMD-like (additive increase multiplicative decrease) admission control policy that is stable irrespective of the total number of nodes and AIMD parameters. The admission policy allows us to establish an event-driven discrete model, triggered by a locally identifiable enabling condition. Subsequently, we propose a decentralized resource allocation strategy via a simple nonlinear state feedback controller, guaranteeing global convergence to a bounded set in finite time. Last, we reveal the connection of these properties with Quality of Service specifications, by calculating local queuing time via a simple formula consistent with Little's Law.
The standard semantics of multi-agent epistemic logic S5 is based on Kripke models whose accessibility relations are reflexive, symmetric and transitive. This one dimensional structure contains implicit higher-dimensional information beyond pairwise interactions, that we formalized as pure simplicial models in a previous work (Information and Computation, 2021). Here we extend the theory to encompass simplicial models that are not necessarily pure. The corresponding class of Kripke models are those where the accessibility relation is symmetric and transitive, but might not be reflexive. Such models correspond to the epistemic logic KB4 . Impure simplicial models arise in situations where two possible worlds may not have the same set of agents. We illustrate it with distributed computing examples of synchronous systems where processes may crash.
Wider coverage and a better solution to a latency reduction in 5G necessitate its combination with multi-access edge computing (MEC) technology. Decentralized deep learning (DDL) such as federated learning and swarm learning as a promising solution to privacy-preserving data processing for millions of smart edge devices, leverages distributed computing of multi-layer neural networks within the networking of local clients, whereas, without disclosing the original local training data. Notably, in industries such as finance and healthcare where sensitive data of transactions and personal medical records is cautiously maintained, DDL can facilitate the collaboration among these institutes to improve the performance of trained models while protecting the data privacy of participating clients. In this survey paper, we demonstrate the technical fundamentals of DDL that benefit many walks of society through decentralized learning. Furthermore, we offer a comprehensive overview of the current state-of-the-art in the field by outlining the challenges of DDL and the most relevant solutions from novel perspectives of communication efficiency and trustworthiness.
Most of our lives are conducted in the cyberspace. The human notion of privacy translates into a cyber notion of privacy on many functions that take place in the cyberspace. This article focuses on three such functions: how to privately retrieve information from cyberspace (privacy in information retrieval), how to privately leverage large-scale distributed/parallel processing (privacy in distributed computing), and how to learn/train machine learning models from private data spread across multiple users (privacy in distributed (federated) learning). The article motivates each privacy setting, describes the problem formulation, summarizes breakthrough results in the history of each problem, and gives recent results and discusses some of the major ideas that emerged in each field. In addition, the cross-cutting techniques and interconnections between the three topics are discussed along with a set of open problems and challenges.