On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

count % of corpus

5 results for “FEMNIST”

Abnormal Client Behavior Detection in Federated Learningcs.LG

This work detects anomalous clients in federated learning to prevent their adverse impacts.

problem Detecting and preventing anomalous client behaviors in federated learning systems.

method Generates low-dimensional surrogates of model weight vectors and uses them for anomaly detection.

result The proposed detection-based approach significantly outperforms conventional defense-based methods.

federated learning anomaly detection client behavior image classification FEMNIST

2019-10-22abs ↗pdf ↗

FedMD: Heterogenous Federated Learning via Model Distillationcs.LG

FedMD combines model distillation for federated learning with private models.

problem Federated learning's limitation in handling private, independently designed models.

method Transfer learning and knowledge distillation.

result Models achieve 20% average accuracy gain through collaboration.

federated learning model distillation private models transfer learning knowledge distillation

2019-10-08abs ↗pdf ↗

Federated Hyperparameter Tuning: Challenges, Baselines, and Connections to Weight-Sharingcs.LG

New method FedEx accelerates federated hyperparameter tuning.

problem Federated hyperparameter tuning challenges in distributed learning.

method FedEx method connecting to weight-sharing, adapted for federated optimization.

result FedEx outperforms natural baselines on various benchmarks.

federated learning hyperparameter tuning weight-sharing FedAvg online convex optimization

2021-06-08abs ↗pdf ↗

Enforcing fairness in private federated learning via the modified method of differential multiplierscs.LG

FPFL mitigates unfairness in private federated learning.

problem Differential privacy degrades model performance on under-represented groups.

method Extends modified method of differential multipliers to private federated learning.

result FPFL reduces unfairness in trained models on private federated learning.

federated learning differential privacy fairness empirical risk minimization group fairness

2021-09-17abs ↗pdf ↗

Communication-Efficient Device Scheduling for Federated Learning Using Stochastic Optimizationcs.LG

Paper optimizes FL communication efficiency with stochastic optimization.

problem Intermittent connectivity and non-i.i.d. data in FL.

method Convergence analysis of non-convex loss functions, stochastic optimization for client selection and power allocation.

result Significant reduction in communication time compared to random participation.

Federated Learning Stochastic Optimization Device Scheduling Communication Efficiency Non-I.I.D. Data

2022-01-19abs ↗pdf ↗