Knowledge Distillation (KD) [13, 3] is a standard tool for transferring information between a machine learning model of lower representational capacity-usually called the student-and a more accurate and powerful teacher model. In the context of classification using neural networks, it is common to consider the student to be a smaller network [2], whereas the teacher is a network that is larger and more computationally-heavy, but also more accurate. Assuming a supervised classification task, distillation consists in training the student to minimize the cross-entropy with respect to the teacher's logits on every given sample, in addition to minimizing the standard cross-entropy loss with respect to the ground truth labels. Since its introduction [3], distillation has been developed and applied in a wide variety of settings, from obtaining compact high-accuracy encodings of model ensembles [13], to boosting the accuracy of compressed models [49, 38, 31], to reinforcement learning [50, 42, 35, 5, 7, 45] and learning with privileged information [51]. Given its apparent simplicity, there has been significant interest in finding explanations for the effectiveness of distillation [2, 13, 37]. For instance, one hypothesis [2, 13] is that the smoothed labels resulting from distillation present the student with a decision surface that is easier to learn than the one presented by the categorical (one-hot) outputs. Another hypothesis [2, 13, 51] starts from the observation that the teacher's outputs have higher entropy than the ground truth labels, and therefore, higher information content. Despite this work, we still have a limited analytical understanding regarding why knowledge distillation is so effective [37].

Federated Learning (FL), a distributed machine learning technique has recently experienced tremendous growth in popularity due to its emphasis on user data privacy. However, the distributed computations of FL can result in constrained communication and drawn-out learning processes, necessitating the client-server communication cost optimization. The ratio of chosen clients and the quantity of local training passes are two hyperparameters that have a significant impact on FL performance. Due to different training preferences across various applications, it can be difficult for FL practitioners to manually select such hyperparameters. In our research paper, we introduce FedAVO, a novel FL algorithm that enhances communication effectiveness by selecting the best hyperparameters leveraging the African Vulture Optimizer (AVO). Our research demonstrates that the communication costs associated with FL operations can be substantially reduced by adopting AVO for FL hyperparameter adjustment. Through extensive evaluations of FedAVO on benchmark datasets, we show that FedAVO achieves significant improvement in terms of model accuracy and communication round, particularly with realistic cases of Non-IID datasets. Our extensive evaluation of the FedAVO algorithm identifies the optimal hyperparameters that are appropriately fitted for the benchmark datasets, eventually increasing global model accuracy by 6% in comparison to the state-of-the-art FL algorithms (such as FedAvg, FedProx, FedPSO, etc.).

Federated Learning (FL) is a collaborative learning framework that enables edge devices to collaboratively learn a global model while keeping raw data locally. Although FL avoids leaking direct information from local datasets, sensitive information can still be inferred from the shared models. To address the privacy issue in FL, differential privacy (DP) mechanisms are leveraged to provide formal privacy guarantee. However, when deploying FL at the wireless edge with over-the-air computation, ensuring client-level DP faces significant challenges. In this paper, we propose a novel wireless FL scheme called private federated edge learning with sparsification (PFELS) to provide client-level DP guarantee with intrinsic channel noise while reducing communication and energy overhead and improving model accuracy. The key idea of PFELS is for each device to first compress its model update and then adaptively design the transmit power of the compressed model update according to the wireless channel status without any artificial noise addition. We provide a privacy analysis for PFELS and prove the convergence of PFELS under general non-convex and non-IID settings. Experimental results show that compared with prior work, PFELS can improve the accuracy with the same DP guarantee and save communication and energy costs simultaneously.

We study the problem of clustering networks whose nodes have imputed or physical positions in a single dimension, for example prestige hierarchies or the similarity dimension of hyperbolic embeddings. Existing algorithms, such as the critical gap method and other greedy strategies, only offer approximate solutions to this problem. Here, we introduce a dynamic programming approach that returns provably optimal solutions in polynomial time -- O(n^2) steps -- for a broad class of clustering objectives. We demonstrate the algorithm through applications to synthetic and empirical networks and show that it outperforms existing heuristics by a significant margin, with a similar execution time.

Networked discrete dynamical systems are often used to model the spread of contagions and decision-making by agents in coordination games. Fixed points of such dynamical systems represent configurations to which the system converges. In the dissemination of undesirable contagions (such as rumors and misinformation), convergence to fixed points with a small number of affected nodes is a desirable goal. Motivated by such considerations, we formulate a novel optimization problem of finding a nontrivial fixed point of the system with the minimum number of affected nodes. We establish that, unless P = NP, there is no polynomial time algorithm for approximating a solution to this problem to within the factor n^1-\epsilon for any constant epsilon > 0. To cope with this computational intractability, we identify several special cases for which the problem can be solved efficiently. Further, we introduce an integer linear program to address the problem for networks of reasonable sizes. For solving the problem on larger networks, we propose a general heuristic framework along with greedy selection methods. Extensive experimental results on real-world networks demonstrate the effectiveness of the proposed heuristics.

This paper investigates new families of compositional optimization problems, called non-smooth weakly-convex finite-sum coupled compositional optimization (NSWC FCCO). There has been a growing interest in FCCO due to its wide-ranging applications in machine learning and AI, as well as its ability to address the shortcomings of stochastic algorithms based on empirical risk minimization. However, current research on FCCO presumes that both the inner and outer functions are smooth, limiting their potential to tackle a more diverse set of problems. Our research expands on this area by examining non-smooth weakly-convex FCCO, where the outer function is weakly convex and non-decreasing, and the inner function is weakly-convex. We analyze a single-loop algorithm and establish its complexity for finding an ϵ-stationary point of the Moreau envelop of the objective function. Additionally, we also extend the algorithm to solving novel non-smooth weakly-convex tri-level finite-sum coupled compositional optimization problems, which feature a nested arrangement of three functions. Lastly, we explore the applications of our algorithms in deep learning for two-way partial AUC maximization and multi-instance two-way partial AUC maximization, using empirical studies to showcase the effectiveness of the proposed algorithms.

This paper studies the joint community detection and phase synchronization problem on the \textit{stochastic block model with relative phase}, where each node is associated with an unknown phase angle. This problem, with a variety of real-world applications, aims to recover the cluster structure and associated phase angles simultaneously. We show this problem exhibits a \textit{``multi-frequency''} structure by closely examining its maximum likelihood estimation (MLE) formulation, whereas existing methods are not originated from this perspective. To this end, two simple yet efficient algorithms that leverage the MLE formulation and benefit from the information across multiple frequencies are proposed. The former is a spectral method based on the novel multi-frequency column-pivoted QR factorization. The factorization applied to the top eigenvectors of the observation matrix provides key information about the cluster structure and associated phase angles. The second approach is an iterative multi-frequency generalized power method, where each iteration updates the estimation in a matrix-multiplication-then-projection manner. Numerical experiments show that our proposed algorithms significantly improve the ability of exactly recovering the cluster structure and the accuracy of the estimated phase angles, compared to state-of-the-art algorithms.

Model-based control requires an accurate model of the system dynamics for precisely and safely controlling the robot in complex and dynamic environments. Moreover, in the presence of variations in the operating conditions, the model should be continuously refined to compensate for dynamics changes. In this paper, we present a self-supervised learning approach that actively models the dynamics of nonlinear robotic systems. We combine offline learning from past experience and online learning from current robot interaction with the unknown environment. These two ingredients enable a highly sample-efficient and adaptive learning process, capable of accurately inferring model dynamics in real-time even in operating regimes that greatly differ from the training distribution. Moreover, we design an uncertainty-aware model predictive controller that is heuristically conditioned to the aleatoric (data) uncertainty of the learned dynamics. This controller actively chooses the optimal control actions that (i) optimize the control performance and (ii) improve the efficiency of online learning sample collection. We demonstrate the effectiveness of our method through a series of challenging real-world experiments using a quadrotor system. Our approach showcases high resilience and generalization capabilities by consistently adapting to unseen flight conditions, while it significantly outperforms classical and adaptive control baselines.

Despite its exceptional soft tissue contrast, Magnetic Resonance Imaging (MRI) faces the challenge of long scanning times compared to other modalities like X-ray radiography. Shortening scanning times is crucial in clinical settings, as it increases patient comfort, decreases examination costs and improves throughput. Recent advances in compressed sensing (CS) and deep learning allow accelerated MRI acquisition by reconstructing high-quality images from undersampled data. While reconstruction algorithms have received most of the focus, designing acquisition trajectories to optimize reconstruction quality remains an open question. This thesis explores two approaches to address this gap in the context of Cartesian MRI. First, we propose two algorithms, lazy LBCS and stochastic LBCS, that significantly improve upon G\"ozc\"u et al.'s greedy learning-based CS (LBCS) approach. These algorithms scale to large, clinically relevant scenarios like multi-coil 3D MR and dynamic MRI, previously inaccessible to LBCS. Additionally, we demonstrate that generative adversarial networks (GANs) can serve as a natural criterion for adaptive sampling by leveraging variance in the measurement domain to guide acquisition. Second, we delve into the underlying structures or assumptions that enable mask design algorithms to perform well in practice. Our experiments reveal that state-of-the-art deep reinforcement learning (RL) approaches, while capable of adaptation and long-horizon planning, offer only marginal improvements over stochastic LBCS, which is neither adaptive nor does long-term planning. Altogether, our findings suggest that stochastic LBCS and similar methods represent promising alternatives to deep RL. They shine in particular by their scalability and computational efficiency and could be key in the deployment of optimized acquisition trajectories in Cartesian MRI.

Reverse engineering deep ReLU networks is a critical problem in understanding the complex behavior and interpretability of neural networks. In this research, we present a novel method for reconstructing deep ReLU networks by leveraging convex optimization techniques and a sampling-based approach. Our method begins by sampling points in the input space and querying the black box model to obtain the corresponding hyperplanes. We then define a convex optimization problem with carefully chosen constraints and conditions to guarantee its convexity. The objective function is designed to minimize the discrepancy between the reconstructed network's output and the target model's output, subject to the constraints. We employ gradient descent to optimize the objective function, incorporating L1 or L2 regularization as needed to encourage sparse or smooth solutions. Our research contributes to the growing body of work on reverse engineering deep ReLU networks and paves the way for new advancements in neural network interpretability and security.