Search Results for author:
Found 27 papers, 12 papers with code
Accelerating Heterogeneous Federated Learning with Closed-form Classifiers
Our findings also indicate that maintaining a fixed classifier aids in stabilizing the training and learning more discriminative features in cross-device settings.
Finding Lottery Tickets in Vision Models via Data-driven Spectral Foresight Pruning
Recent advances in neural network pruning have shown how it is possible to reduce the computational costs and memory demands of deep learning models before training.
PEM: Prototype-based Efficient MaskFormer for Image Segmentation
To fill this gap, we propose Prototype-based Efficient MaskFormer (PEM), an efficient transformer-based architecture that can operate in multiple segmentation tasks.
Communication-Efficient Heterogeneous Federated Learning with Generalized Heavy-Ball Momentum
Federated Learning (FL) has emerged as the state-of-the-art approach for learning from decentralized data in privacy-constrained scenarios.
From Charts to Atlas: Merging Latent Spaces into One
Models trained on semantically related datasets and tasks exhibit comparable inter-sample relations within their latent spaces.
Window-based Model Averaging Improves Generalization in Heterogeneous Federated Learning
To address these issues and improve the robustness and generalization capabilities of the global model, we propose WIMA (Window-based Model Averaging).
FedDrive v2: an Analysis of the Impact of Label Skewness in Federated Semantic Segmentation for Autonomous Driving
We propose FedDrive v2, an extension of the Federated Learning benchmark for Semantic Segmentation in Autonomous Driving.
Schedule-Robust Online Continual Learning
A continual learning (CL) algorithm learns from a non-stationary data stream.
Learning Across Domains and Devices: Style-Driven Source-Free Domain Adaptation in Clustered Federated Learning
Federated Learning (FL) has recently emerged as a possible way to tackle the domain shift in real-world Semantic Segmentation (SS) without compromising the private nature of the collected data.
A Marriage between Adversarial Team Games and 2-player Games: Enabling Abstractions, No-regret Learning, and Subgame Solving
This work shows that we can recover from this weakness by bridging the gap between sequential adversarial team games and 2-player games.
Fault-Aware Design and Training to Enhance DNNs Reliability with Zero-Overhead
Deep Neural Networks (DNNs) enable a wide series of technological advancements, ranging from clinical imaging, to predictive industrial maintenance and autonomous driving.
Improving Generalization in Federated Learning by Seeking Flat Minima
Models trained in federated settings often suffer from degraded performances and fail at generalizing, especially when facing heterogeneous scenarios.
FedDrive: Generalizing Federated Learning to Semantic Segmentation in Autonomous Driving
For similar reasons, Federated Learning has been recently introduced as a new machine learning paradigm aiming to learn a global model while preserving privacy and leveraging data on millions of remote devices.
Speeding up Heterogeneous Federated Learning with Sequentially Trained Superclients
data severely impairs both the performance of the trained neural network and its convergence rate, increasing the number of communication rounds requested to reach a performance comparable to that of the centralized scenario.
Public Information Representation for Adversarial Team Games
Interestingly, we show that our game is more expressive than the original extensive-form game as any state/action abstraction of the extensive-form game can be captured by our representation, while the reverse does not hold.
Incremental Learning in Semantic Segmentation from Image Labels
As opposed to existing approaches, that need to generate pseudo-labels offline, we use an auxiliary classifier, trained with image-level labels and regularized by the segmentation model, to obtain pseudo-supervision online and update the model incrementally.
Cluster-driven Graph Federated Learning over Multiple Domains
Clustering may reduce heterogeneity by identifying the domains, but it deprives each cluster model of the data and supervision of others.
DA4Event: towards bridging the Sim-to-Real Gap for Event Cameras using Domain Adaptation
Event cameras are novel bio-inspired sensors, which asynchronously capture pixel-level intensity changes in the form of "events".
Multi-Agent Coordination in Adversarial Environments through Signal Mediated Strategies
Team members can coordinate their strategies before the beginning of the game, but are unable to communicate during the playing phase of the game.
A Differentiable Recurrent Surface for Asynchronous Event-Based Data
Dynamic Vision Sensors (DVSs) asynchronously stream events in correspondence of pixels subject to brightness changes.
Coordination in Adversarial Sequential Team Games via Multi-Agent Deep Reinforcement Learning
Many real-world applications involve teams of agents that have to coordinate their actions to reach a common goal against potential adversaries.
Attention Mechanisms for Object Recognition with Event-Based Cameras
Event-based cameras are neuromorphic sensors capable of efficiently encoding visual information in the form of sparse sequences of events.
ReConvNet: Video Object Segmentation with Spatio-Temporal Features Modulation
We introduce ReConvNet, a recurrent convolutional architecture for semi-supervised video object segmentation that is able to fast adapt its features to focus on any specific object of interest at inference time.
Asynchronous Convolutional Networks for Object Detection in Neuromorphic Cameras
Event-based cameras, also known as neuromorphic cameras, are bioinspired sensors able to perceive changes in the scene at high frequency with low power consumption.
NAIS-Net: Stable Deep Networks from Non-Autonomous Differential Equations
This paper introduces Non-Autonomous Input-Output Stable Network(NAIS-Net), a very deep architecture where each stacked processing block is derived from a time-invariant non-autonomous dynamical system.
Multi-View Stereo with Single-View Semantic Mesh Refinement
In this paper we propose a novel method to refine both the geometry and the semantic labeling of a given mesh.
ReSeg: A Recurrent Neural Network-based Model for Semantic Segmentation
Moreover, ReNet layers are stacked on top of pre-trained convolutional layers, benefiting from generic local features.
Ranked #21 on
Semantic Segmentation
on CamVid
