New method lowers spherical perceptron capacity using fully lifted random duality theory.
problem Tackles the negative spherical perceptron capacity, a long-standing open problem.
method Develops fully lifted random duality theory (fl RDT) to characterize capacity.
result Shows remarkable closed-form analytical relations for practical capacity values.
Wide hidden layer TCM nets capacity analyzed using RDT and fl RDT.
problem Capacity analysis of wide hidden layer TCM nets.
method Employed Fully Lifted Random Duality Theory (fl RDT) for capacity characterization.
result Explicit, closed form capacity characterizations for a generic class of hidden layer activations.
New insights into binary perceptron reveal phase transitions and algorithmic thresholds.
problem Understanding the statistical-computational gap in binary perceptron models.
method Application of fully lifted random duality theory (fl RDT) to uncover structural changes.
result Numerical estimates of constraint density thresholds align with theoretical predictions.
Study potential computational gaps in symmetric binary perceptrons using fl-RDT.
problem Potential statistical-computational gaps in symmetric binary perceptrons.
method Parametric utilization of fully lifted random duality theory (fl-RDT).
result Observation of a computational gap SCG=αc−αa in SBP. Improved neural network capacity analysis using simplified RDT.
problem Analyzing the memorization capabilities of sign perceptron neural networks.
method Developed a simplified, partially lifted Random Duality Theory (fl RDT) approach.
result Concrete capacity bounds universally improve over previous best known ones.
Study on theoretical limits of ℓ0 sparse-regression algorithms using Fl RDT.
problem Understanding the performance limits of ℓ0 norm based optimization algorithms in compressed sensing and sparse regression. method Utilized Fully lifted random duality theory (Fl RDT) to analyze the maximum-likelihood (ML) decoding performance.
result Uncovered phase-transition (PT) and descending ℓ0 (dℓ0) curves that separate successful and unsuccessful algorithm performance. Study binary perceptrons' capacity using random duality theory.
problem Characterize the capacity of binary perceptrons with general thresholds.
method Utilized fully lifted random duality theory (fl RDT) to characterize the capacity.
result Characterizations match replica symmetry breaking predictions and uncover the capacity for zero-threshold scenario.
This paper connects ultrametric overlap gap properties to parametric RDT for symmetric binary perceptrons.
problem Characterizing statistical computational gaps in symmetric binary perceptrons.
method Developed an analytical union-bounding program to rigorously upper-bound constraint densities of ultrametric overlap gap properties.
result Obtained tightest bounds at the first two levels of ultrametric overlap gap properties, closely approaching parametric RDT estimates.
The study calculates the injectivity capacity of ReLU networks using a novel mathematical approach.
problem Determining the injectivity capacity of ReLU networks layers.
method Employing fully lifted random duality theory (fl RDT) to handle the ℓ0 spherical perceptron and implicitly the ReLU layers injectivity. result The lifting mechanism converges remarkably fast with relative corrections not exceeding 0.1%.
New method finds rare dense clusters in asymmetric binary perceptrons, resolving algorithmic hardness.
problem Resolving algorithmic hardness in asymmetric binary perceptrons.
method Fully lifted random duality theory (fl RDT) and large deviation upgrade (sfl LD RDT).
result Local entropy breaks down for constraint densities in (0.77, 0.78) interval, matching current solver limits.
The study revisits Hopfield's associative memory model and calculates its capacity for two specific pattern basins.
problem Determining the capacity of a Hebbian-Hopfield network for storing binary patterns.
method Using fully lifted random duality theory and numerical analysis, the study calculates the capacity for two specific pattern basins.
result Explicit characterizations of the capacity for the AGS and NLT pattern basins, with remarkable fast lifting convergence.
New analysis shows capacity of treelike neural networks with various activations.
problem Analyzing the capacity of treelike neural networks with diverse activations.
method Utilized Random Duality Theory and its partially lifted version to handle various activations.
result The capacity of treelike neural networks decreases for large network width but converges to a constant value.
Study precise sample covariance error for Gaussian centered data.
problem Precise characterization of sample covariance error for Gaussian data.
method Developed a Random Duality Theory (RDT) framework to determine upper and lower bounds.
result Upper and lower bounds match in large-dimensional contexts, matching the spectral norm's limiting value.
CLuP achieves near optimal ground state energies for positive and negative Hopfield models.
problem Finding near optimal ground state energies for positive and negative Hopfield models.
method Controlled Loosening-up (CLuP) algorithm with fully lifted random duality theory (fl RDT).
result Achieves ground state free energies of 1.77 and 0.33 for positive and negative Hopfield models respectively. This work adapts RDT for mental program construction, showing benefits and costs.
problem Applying RDT to mental programs with trade-offs between description length, error, and computational costs.
method Proposed a three-way trade-off and used simulations and partial information decomposition.
result Constructing a shared program library provides global benefits but is sensitive to curricula.
Optimizes convergence time of federated learning over wireless networks.
problem Limited resource blocks in wireless networks affect federated learning convergence time and performance.
method Formulates an optimization problem to minimize convergence time while optimizing performance, proposes a probabilistic user selection scheme and uses ANNs for estimation.
result Improves convergence time and performance of federated learning over wireless networks.
In this paper, the problem of training federated learning (FL) algorithms over a realistic wireless network is studied. In particular, in the considered model, wireless users execute an FL algorithm while training their local FL models using their own data and transmitting the trained local FL models to a base station …
This work bridges federated learning and contextual bandits, enhancing FL's utility.
problem Limited use of federated learning in contextual bandits despite its potential.
method Proposes FedIGW, a novel federated contextual bandits design that leverages regression-based algorithms and integrates various FL components.
result FedIGW better harnesses FL innovations and provides flexible, modular, and seamless integration of FL elements.
Group personalization improves FL performance in heterogeneous client data.
problem Mitigating client drift in federated learning with heterogeneous data.
method Fine-tuning a global FL model over homogeneous groups of clients, then personalizing each group's model.
result The proposed method achieves superior personalization performance compared to other FL approaches.
This paper analyzes convergence of FL for neural networks using NTK.
problem Theoretical guarantees of FL for neural networks with explicit forms and multi-step updates are unexplored.
method FL-NTK framework for federated learning of ReLU neural networks trained by gradient descent.
result FL-NTK converges to a global-optimal solution at a linear rate with proper learning parameters.
Efficient Bayesian FL method improves predictive accuracy and uncertainty estimates.
problem Federated Learning with model and predictive uncertainty and personalization.
method Second-order optimization approach for Bayesian FL.
result Improved predictive accuracies and uncertainty estimates.
Flower framework simplifies federated learning experiments on edge devices.
problem Realistic implementation of Federated Learning on edge devices is challenging.
method Developed a comprehensive federated learning framework, Flower, supporting large-scale experiments on heterogeneous devices.
result Flower enables federated learning experiments with up to 15M client size using only two high-end GPUs.
Semi-Federated Learning clusters clients for efficient model training.
problem Efficiency and data distribution challenges in Federated Learning.
method Local clustering and in-cluster training with a sequential training manner.
result Semi-Federated Learning reduces communication costs and improves robustness to Non-IID data.
Study quantifies impacts of heterogeneity in FL on smartphone data.
problem Heterogeneity in FL devices causes performance degradation.
method Collected 136k smartphone data, built heterogeneity-aware FL platform, conducted extensive experiments.
result Heterogeneity causes up to 9.2% accuracy drop and 2.32x training time increase.
FL's early training phase significantly impacts final test accuracy.
problem Understanding how early phases affect FL's final test accuracy.
method Generalized Fisher Information Matrix (FedFIM) to FL.
result FL exhibits critical learning periods where small errors can have large impacts.
Flashback Learning balances model stability and plasticity in continual learning.
problem Balancing model stability and plasticity in continual learning.
method Flashback Learning (FL) uses a bidirectional regularization approach to balance stability and plasticity.
result FL improves model accuracy by up to 4.91% in Class-Incremental and 3.51% in Task-Incremental settings.
New bounds on neural network capacity for treelike sign perceptrons using RDT.
problem Determining the capacity of treelike sign perceptrons neural networks.
method Random Duality Theory (RDT) to establish upper bounds.
result Mathematically rigorous bounds on network capacity for any number of neurons.
Survey on threats to federated learning models.
problem Vulnerabilities in federated learning protocols.
method Taxonomy of threat models and attacks.
result Important future research directions.
A new asynchronous method for vertical federated learning improves privacy and efficiency.
problem Solving vertical federated learning in an asynchronous manner with privacy and efficiency.
method A simple FL method that allows clients to run stochastic gradient algorithms asynchronously with a new perturbed local embedding technique.
result The method improves privacy and communication efficiency compared to centralized and synchronous FL methods.
FL+HC improves federated learning on non-iid data by clustering local updates.
problem FL struggles with non-iid data, leading to suboptimal models.
method Introduce hierarchical clustering to separate and train clusters of clients independently.
result FL+HC converges faster and achieves higher accuracy than standard FL.
NAC-FL optimizes model updates in FL systems by adapting compression to network congestion.
problem Federated Learning systems face congestion and delays in data exchanges.
method NAC-FL dynamically adjusts client compression based on network congestion.
result NAC-FL reduces training time and achieves robust performance improvements.
Optimal feature learning strength improves generalization in deep networks.
problem Understanding how feature learning strength affects generalization in practical settings.
method Empirical studies and theoretical analysis of gradient flow dynamics in two-layer ReLU nets.
result Optimal feature learning strength yields substantial generalization gains, contrary to the prevailing intuition.
Paper proposes SCALLION and SCAFCOM for compressed FL with reduced communication.
problem Reducing communication overhead in Federated Learning with data heterogeneity and partial participation.
method Revisit and simplify stochastic controlled averaging, proposing SCALLION and SCAFCOM for unbiased and biased compression.
result SCALLION and SCAFCOM outperform existing methods in communication and computation complexities.
Study precise estimators for correlated data using RDT.
problem Analyzing estimators in correlated linear regression models.
method Utilized Random Duality Theory to characterize prediction risk.
result Precise closed form characterizations of estimators' risk.
We connect Causal inference and low-rank recovery via RDT and free probability theory.
problem Determining the applicability of causal inference via low-rank recovery.
method Random Duality Theory, free probability theory, and mathematical rigor.
result Exact closed-form worst case phase transitions for causal inference.
New algorithm nearly achieves ground state free energy of SK model.
problem Determining the ground state free energy of the SK model.
method Controlled Loosening-up (CLuP) algorithm applied to SK models.
result Achieves ground state free energy of ~0.76 for n in the thousands.
FL-Sailer enables federated learning for scATAC-seq data, reducing dimensionality and noise.
problem Privacy-preserving federated learning for ultra-high dimensional, sparse, and heterogeneous scATAC-seq data.
method FL-Sailer integrates adaptive leverage score sampling and an invariant VAE architecture.
result FL-Sailer converges to an approximate solution with bounded error, surpassing centralized methods.
FedML aims to improve FL research by providing a library and benchmark.
problem Inconsistent FL algorithm development and performance comparison.
method FedML offers an open research library and benchmark supporting diverse computing paradigms and flexible API design.
result FedML facilitates fair algorithm comparison and development in federated learning.
FOCUS addresses label quality disparity in FL for healthcare applications.
problem Label quality disparity in federated learning for healthcare applications.
method FOCUS maintains a small set of benchmark samples and computes the mutual cross-entropy between local and benchmark datasets to quantify label credibility. It then adjusts client weights based on credibility values.
result FOCUS effectively reduces the impact of noisy labels from clients, improving model performance.
This work analyzes generalization in federated learning using information theory.
problem Generalization performance in federated learning is less explored compared to centralized learning.
method The work applies an information-theoretic analysis via the conditional mutual information (CMI) framework to study federated learning's two-level generalization.
result The work derives multiple CMI-based bounds, including hypothesis-based CMI bounds and fast-rate evaluated CMI bounds, which improve convergence rates for specific model aggregation strategies and structured loss functions.
Enhances OTA FL algorithms by defining inverse feasibility for linear models.
problem Improving security and privacy in over-the-air federated learning.
method Defines inverse feasibility as an upper bound on condition number, analyzes existing model, proposes new model.
result Proposes a new OTA FL model with enhanced characteristics.
FLeet improves online FL for mobile apps with better performance and privacy.
problem Federated Learning's offline nature limits its applicability for online updates.
method Combines staleness awareness and performance prediction with adaptive learning.
result 2.3x quality boost with minimal battery consumption.
SIGMA prior enables federated learning for non-factorizable models.
problem Current FL methods assume conditional independence, limiting applicability to non-factorizable models.
method SIGMA prior approximates deep generative model to induce conditional independence structure.
result SIGMA prior expands FL applicability to fields requiring modeling dependencies.
Multi-model FL improves performance without sharing data.
problem Training multiple models in a federated setting without data sharing.
method Proposed two variants of FedAvg for multi-model FL with provable convergence guarantees.
result Multi-model FL can have better performance than training each model separately.
Local adaptation improves federated learning models.
problem Improving accuracy of federated learning models on non-iid data.
method Local adaptation techniques (fine-tuning, multi-task learning, knowledge distillation).
result Participants benefit from local adaptation, improving federated model accuracy.
Federated learning (FL) allows model training from local data collected by edge/mobile devices while preserving data privacy, which has wide applicability to image and vision applications. A challenge is that client devices in FL usually have much more limited computation and communication resources compared to servers…
This thesis tackles FL challenges with new methods and algorithms.
problem Privacy-preserving machine learning with decentralized data.
method Compression, client selection, and heterogeneity handling.
result Practical FL solutions with mathematically rigorous guarantees.
LC-FL uses generative models to reduce communication costs in federated learning.
problem High communication costs and strict model homogeneity in federated learning.
method LC-FL employs generative models to transmit data and aggregate models.
result LC-FL reduces communication costs and supports heterogeneous models.