Logo: to the web site of Uppsala University

The rise of Big Data has catalyzed numerous advanced data-driven methods, while simultaneously posing significant challenges in data management. This thesis aims to address two fundamental aspects of data management–storage management and information extraction–by leveraging machine learning (ML) techniques. In particular, we focus on two research topics: Storage Hierarchy, which explores hierarchical storage management (HSM) in multi-tiered storage systems; and Information Hierarchy, which targets the extraction of intrinsic data hierarchies from raw data.

We begin by introducing the key stages of data life cycle and their associated challenges in the Big Data era, alongside a review of machine learning foundations and their potentials for addressing these challenges. Subsequently, we present the Storage Hierarchy project, which is detailed across Paper I, II, and III. In these works, we develop automated, adaptive, and efficient HSM approaches using reinforcement learning (RL). In Paper I we introduce the HSM-RL framework for managing file-level data migration in hierarchical storage system (HSS). It leverages RL to optimize file placement and temporal difference learning for real-time adaptability. Paper II extends this work to complex real–world scenarios using scientific datasets, exploring the framework’s flexibility, scalability, and effectiveness. Moving to finer granularity, Paper III presents ReStore, an RL-based page-level data migration approach that incorporates the unique characteristics of modern Solid-State Drives (SSDs), such as read/write asymmetry and parallelism.

The Information Hierarchy project focuses on autonomous extraction of implicit data hierarchies from raw, unlabeled data. Presented in Paper IV, we propose InfoHier, a framework that integrates self-supervised learning (SSL) with hierarchical clustering (HC) to uncover latent data representations and hierarchical structures. By jointly training SSL and HC through a dynamic balancing loss, InfoHier ensure that the HC results align with the intrinsic data hierarchy. This method facilitates meaningful and structured information extraction and retrieval. 

Collectively, the Storage Hierarchy and Information Hierarchy projects advance intelligent data management by enabling efficient storage solutions and autonomous information extraction. These contributions pave the foundation for next generation data management systems, addressing the challenges of Big Data with adaptive and scalable solutions.

Federated learning is a distributed and privacy-preserving approach to train a statistical model collaboratively from decentralized data held by different parties. However, when the datasets are not independent and identically distributed, models trained by naive federated algorithms may be biased towards certain participants, and model performance across participants is non-uniform. This is known as the fairness problem in federated learning. In this paper, we formulate fairness-controlled federated learning as a dynamical multi-objective optimization problem to ensure the fairness and convergence with theoretical guarantee. To solve the problem efficiently, we study the convergence and bias of Adam as the server optimizer in federated learning, and propose Adaptive Federated Adam ( AdaFedAdam ) to accelerate fair federated learning with alleviated bias. We validated the effectiveness, Pareto optimality and robustness of AdaFedAdam with numerical experiments and show that AdaFedAdam outperforms existing algorithms, providing better convergence and fairness properties of the federated scheme.

In the present era of big data, the challenges of data management have grown significantly. One crucial aspect is the management of data storage. As data volumes continue to expand, effective storage management becomes increasingly essential. Meanwhile, evolving hardware technologies offer various storage options, ranging from HDDs to SSDs and NVRAMs. To this end, hierarchical (multi-tier) storage systems (HSS) have emerged as a solution, organizing different storage devices hierarchically to provide various storage options. However, managing multiple storage tiers and their data, while optimizing performance and cost-efficiency, is extremely complex. In this paper, we discuss the challenges in the management of hierarchical storage system. We summarise our previous contributions on tackling these challenges, including the proposal of a reinforcement learning (RL) based data migration policy and the design of an autonomous hierarchical storage management framework HSM-RL. We also present the applications of HSM-RL in scientific data management to demonstrate its adaptability and scalability. Finally, we conclude our work to date and outline the future research plans.

Federated learning (FL) facilitates distributed training across different IoT and edge devices, safeguarding the privacy of their data. The inherent distributed structure of FL introduces vulnerabilities, especially from adversarial devices aiming to skew local updates to their advantage. Despite the plethora of research focusing on Byzantine-resilient FL, the academic community has yet to establish a comprehensive benchmark suite, pivotal for impartial assessment and comparison of different techniques. This paper presents Blades, a scalable, extensible, and easily configurable benchmark suite that supports researchers and developers in efficiently implementing and validating novel strategies against baseline algorithms in Byzantine-resilient FL. Blades contains built-in implementations of representative attack and defense strategies and offers a user-friendly interface that seamlessly integrates new ideas. Using Blades, we re-evaluate representative attacks and defenses on wide-ranging experimental configurations (approximately 1,500 trials in total). Through our extensive experiments, we gained new insights into FL robustness and highlighted previously overlooked limitations due to the absence of thorough evaluations and comparisons of baselines under various attack settings. We maintain the source code and documents at https://github.com/lishenghui/blades.

In many areas of data-driven science, large datasets are generated where the individual data objects are images, matrices, or otherwise have a clear structure. However, these objects can be information-sparse, and a challenge is to efficiently find and work with the most interesting data as early as possible in an analysis pipeline. We have recently proposed a new model for big data management where the internal structure and information of the data are associated with each data object (as opposed to simple metadata). There is then an opportunity for comprehensive data management solutions to account for data-specific internal structure as well as access patterns. In this article, we explore this idea together with our recently proposed hierarchical storage management framework that uses reinforcement learning (RL) for autonomous and dynamic data placement in different tiers in a storage hierarchy. Our case-study is based on four scientific datasets: Protein translocation microscopy images, Airfoil angle of attack meshes, 1000 Genomes sequences, and Phenotypic screening images. The presented results highlight that our framework is optimal and can quickly adapt to new data access requirements. It overall reduces the data processing time, and the proposed autonomous data placement is superior compared to any static or semi-static data placement policies.

Federated learning (FL) facilitates distributed training across different IoT and edge devices, safeguarding the privacy of their data. The inherently distributed nature of FL introduces vulnerabilities, especially from adversarial devices aiming to skew local updates to their desire. Despite the plethora of research focusing on Byzantine-resilient FL, the academic conununity has yet to establish a comprehensive benchmark suite, pivotal for the assessment and comparison of different techniques. This demonstration presents Blades, a scalable, extensible, and easily configurable benchmark suite that supports researchers and developers in efficiently implementing and validating strategies against baseline algorithms in Byzantine-resilient FL.

With the rapid development of big data and cloud computing, data management has become increasingly challenging. Over the years, a number of frameworks for data management have become available. Most of them are highly efficient, but ultimately create data silos. It becomes difficult to move and work coherently with data as new requirements emerge. A possible solution is to use an intelligent hierarchical (multi-tier) storage system (HSS). A HSS is a meta solution that consists of different storage frameworks organized as a jointly constructed storage pool. A built-in data migration policy that determines the optimal placement of the datasets in the hierarchy is essential. Placement decisions is a non-trivial task since it should be made according to the characteristics of the dataset, the tier status in a hierarchy, and access patterns. This paper presents an open-source hierarchical storage framework with a dynamic migration policy based on reinforcement learning (RL). We present a mathematical model, a software architecture, and implementations based on both simulations and a live cloud-based environment. We compare the proposed RL-based strategy to a baseline of three rule-based policies, showing that the RL-based policy achieves significantly higher efficiency and optimal data distribution in different scenarios.

With the rapid development of big data and cloud computing, data management has become increasingly challenging. A possible solution is to use an intelligent hierarchical (multi-tier) storage system (HSS). An HSS is a meta solution that consists of different storage frameworks organized as a jointly constructed storage pool. A built-in data migration policy that determines the optimal placement of the datasets in the hierarchy is essential. Placement decisions are a non-trivial task since they should be made according to the characteristics of the dataset, the tier status in a hierarchy, and access patterns. This paper presents an open-source hierarchical storage framework with a dynamic migration policy based on reinforcement learning (RL).

Analyzing large-scale datasets, especially involving complex and high-dimensional data like images, is particularly challenging. While self-supervised learning (SSL) has proven effective for learning representations from unlabeled data, it typically focuses on flat, non-hierarchical structures, missing the multi-level relationships present in many real-world datasets. Hierarchical clustering (HC) can uncover these relationships by organizing data into a tree-like structure, but it often relies on rigid similarity metrics that struggle to capture the complexity of diverse data types. To address these we envision InfoHier, a framework that combines SSL with HC to jointly learn robust latent representations and hierarchical structures. This approach leverages SSL to provide adaptive representations, enhancing HC's ability to capture complex patterns. Simultaneously, it integrates HC loss to refine SSL training, resulting in representations that are more attuned to the underlying information hierarchy. InfoHier has the potential to improve the expressiveness and performance of both clustering and representation learning, offering significant benefits for data analysis, management, and information retrieval.