Search Results for author: Marcello Pelillo

Found 49 papers, 21 papers with code

Minimizing Energy Consumption of Deep Learning Models by Energy-Aware Training

no code implementations1 Jul 2023 Dario Lazzaro, Antonio Emanuele Cinà, Maura Pintor, Ambra Demontis, Battista Biggio, Fabio Roli, Marcello Pelillo

Deep learning models undergo a significant increase in the number of parameters they possess, leading to the execution of a larger number of operations during inference.

Reassembling Broken Objects using Breaking Curves

1 code implementation5 Jun 2023 Ali Alagrami, Luca Palmieri, Sinem Aslan, Marcello Pelillo, Sebastiano Vascon

Results show that our solution performs well in reassembling different kinds of broken objects.


Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoning

no code implementations4 May 2022 Antonio Emanuele Cinà, Kathrin Grosse, Ambra Demontis, Sebastiano Vascon, Werner Zellinger, Bernhard A. Moser, Alina Oprea, Battista Biggio, Marcello Pelillo, Fabio Roli

In this survey, we provide a comprehensive systematization of poisoning attacks and defenses in machine learning, reviewing more than 100 papers published in the field in the last 15 years.

BIG-bench Machine Learning Data Poisoning

Machine Learning Security against Data Poisoning: Are We There Yet?

1 code implementation12 Apr 2022 Antonio Emanuele Cinà, Kathrin Grosse, Ambra Demontis, Battista Biggio, Fabio Roli, Marcello Pelillo

The recent success of machine learning (ML) has been fueled by the increasing availability of computing power and large amounts of data in many different applications.

BIG-bench Machine Learning Data Poisoning

The Group Loss++: A deeper look into group loss for deep metric learning

no code implementations4 Apr 2022 Ismail Elezi, Jenny Seidenschwarz, Laurin Wagner, Sebastiano Vascon, Alessandro Torcinovich, Marcello Pelillo, Laura Leal-Taixe

Deep metric learning has yielded impressive results in tasks such as clustering and image retrieval by leveraging neural networks to obtain highly discriminative feature embeddings, which can be used to group samples into different classes.

Clustering Image Retrieval +3

Relaxation Labeling Meets GANs: Solving Jigsaw Puzzles with Missing Borders

no code implementations28 Mar 2022 Marina Khoroshiltseva, Arianna Traviglia, Marcello Pelillo, Sebastiano Vascon

This paper proposes JiGAN, a GAN-based method for solving Jigsaw puzzles with eroded or missing borders.

Energy-Latency Attacks via Sponge Poisoning

2 code implementations14 Mar 2022 Antonio Emanuele Cinà, Ambra Demontis, Battista Biggio, Fabio Roli, Marcello Pelillo

Sponge examples are test-time inputs carefully optimized to increase energy consumption and latency of neural networks when deployed on hardware accelerators.

Federated Learning Test

Can machines learn to see without visual databases?

no code implementations12 Oct 2021 Alessandro Betti, Marco Gori, Stefano Melacci, Marcello Pelillo, Fabio Roli

This paper sustains the position that the time has come for thinking of learning machines that conquer visual skills in a truly human-like context, where a few human-like object supervisions are given by vocal interactions and pointing aids only.


Backdoor Learning Curves: Explaining Backdoor Poisoning Beyond Influence Functions

1 code implementation14 Jun 2021 Antonio Emanuele Cinà, Kathrin Grosse, Sebastiano Vascon, Ambra Demontis, Battista Biggio, Fabio Roli, Marcello Pelillo

Backdoor attacks inject poisoning samples during training, with the goal of forcing a machine learning model to output an attacker-chosen class when presented a specific trigger at test time.

BIG-bench Machine Learning Incremental Learning +1

The Hammer and the Nut: Is Bilevel Optimization Really Needed to Poison Linear Classifiers?

1 code implementation23 Mar 2021 Antonio Emanuele Cinà, Sebastiano Vascon, Ambra Demontis, Battista Biggio, Fabio Roli, Marcello Pelillo

One of the most concerning threats for modern AI systems is data poisoning, where the attacker injects maliciously crafted training data to corrupt the system's behavior at test time.

Bilevel Optimization Data Poisoning

Transductive Visual Verb Sense Disambiguation

1 code implementation20 Dec 2020 Sebastiano Vascon, Sinem Aslan, Gianluca Bigaglia, Lorenzo Giudice, Marcello Pelillo

Verb Sense Disambiguation is a well-known task in NLP, the aim is to find the correct sense of a verb in a sentence.


DSLib: An open source library for the dominant set clustering method

1 code implementation15 Oct 2020 Sebastiano Vascon, Samuel Rota Bulò, Vittorio Murino, Marcello Pelillo

This package provides an implementation of the original DS clustering algorithm since no code has been officially released yet, together with a still growing collection of methods and variants related to it.

Clustering Graph Matching

Semantic Change Detection with Asymmetric Siamese Networks

1 code implementation12 Oct 2020 Kunping Yang, Gui-Song Xia, Zicheng Liu, Bo Du, Wen Yang, Marcello Pelillo, Liangpei Zhang

Given two multi-temporal aerial images, semantic change detection aims to locate the land-cover variations and identify their change types with pixel-wise boundaries.

Change Detection Management

A black-box adversarial attack for poisoning clustering

1 code implementation9 Sep 2020 Antonio Emanuele Cinà, Alessandro Torcinovich, Marcello Pelillo

In an attempt to fill this gap, in this work, we propose a black-box adversarial attack for crafting adversarial samples to test the robustness of clustering algorithms.

Adversarial Attack Clustering +1

Efficient Tensor Kernel methods for sparse regression

no code implementations23 Mar 2020 Feliks Hibraj, Marcello Pelillo, Saverio Salzo, Massimiliano Pontil

Second, we use a Nystrom-type subsampling approach, which allows for a training phase with a smaller number of data points, so to reduce the computational cost.


The Group Loss for Deep Metric Learning

2 code implementations ECCV 2020 Ismail Elezi, Sebastiano Vascon, Alessandro Torcinovich, Marcello Pelillo, Laura Leal-Taixe

Deep metric learning has yielded impressive results in tasks such as clustering and image retrieval by leveraging neural networks to obtain highly discriminative feature embeddings, which can be used to group samples into different classes.

Ranked #19 on Metric Learning on CUB-200-2011 (using extra training data)

Clustering Image Retrieval +2

Weakly Supervised Semantic Segmentation Using Constrained Dominant Sets

no code implementations20 Sep 2019 Sinem Aslan, Marcello Pelillo

The availability of large-scale data sets is an essential pre-requisite for deep learning based semantic segmentation schemes.

Segmentation Weakly supervised Semantic Segmentation +1

Dominant Set Clustering and Pooling for Multi-View 3D Object Recognition

1 code implementation4 Jun 2019 Chu Wang, Marcello Pelillo, Kaleem Siddiqi

We improve upon these methods by introducing a view clustering and pooling layer based on dominant sets.

3D Object Recognition Clustering +1

Unsupervised Domain Adaptation using Graph Transduction Games

no code implementations6 May 2019 Sebastiano Vascon, Sinem Aslan, Alessandro Torcinovich, Twan van Laarhoven, Elena Marchiori, Marcello Pelillo

Unsupervised domain adaptation (UDA) amounts to assigning class labels to the unlabeled instances of a dataset from a target domain, using labeled instances of a dataset from a related source domain.

Object Recognition Unsupervised Domain Adaptation

Deep Constrained Dominant Sets for Person Re-identification

1 code implementation ICCV 2019 Leulseged Tesfaye Alemu, Marcello Pelillo, Mubarak Shah

By optimizing the constrained clustering in an end-to-end manner, we naturally leverage the contextual knowledge of a set of images corresponding to the given person-images.

Ranked #2 on Person Re-Identification on CUHK03 (Rank-5 metric)

Constrained Clustering Image Retrieval +2

A Functional Representation for Graph Matching

1 code implementation16 Jan 2019 Fu-Dong Wang, Gui-Song Xia, Nan Xue, Yi-Peng Zhang, Marcello Pelillo

In this paper, we present a functional representation for graph matching (FRGM) that aims to provide more geometric insights on the problem and reduce the space and time complexities of corresponding algorithms.

Graph Matching

Is Data Clustering in Adversarial Settings Secure?

no code implementations25 Nov 2018 Battista Biggio, Ignazio Pillai, Samuel Rota Bulò, Davide Ariu, Marcello Pelillo, Fabio Roli

In this work we propose a general framework that allows one to identify potential attacks against clustering algorithms, and to evaluate their impact, by making specific assumptions on the adversary's goal, knowledge of the attacked system, and capabilities of manipulating the input data.


Ancient Coin Classification Using Graph Transduction Games

no code implementations2 Oct 2018 Sinem Aslan, Sebastiano Vascon, Marcello Pelillo

Experiments are conducted on the only publicly available dataset which is composed of 180 images of 60 types of Roman coins.

Classification General Classification +1

Multi-feature Fusion for Image Retrieval Using Constrained Dominant Sets

no code implementations15 Aug 2018 Leulseged Tesfaye Alemu, Marcello Pelillo

In this paper, we propose a computationally efficient approach to fuse several hand-crafted and deep features, based on the probabilistic distribution of a given membership score of a constrained cluster in an unsupervised manner.

Image Retrieval Retrieval

A Graph Transduction Game for Multi-target Tracking

no code implementations12 Jun 2018 Tewodros Mulugeta Dagnew, Dalia Coppi, Marcello Pelillo, Rita Cucchiara

Semi-supervised learning is a popular class of techniques to learn from labeled and unlabeled data.

Multiple People Tracking

Transductive Label Augmentation for Improved Deep Network Learning

no code implementations26 May 2018 Ismail Elezi, Alessandro Torcinovich, Sebastiano Vascon, Marcello Pelillo

Motivated by the observation that in virtually all application domains it is easy to obtain unlabeled data, in this paper we take a different perspective and propose a \emph{label augmentation} approach.

Data Augmentation General Classification +2

Speaker Clustering Using Dominant Sets

1 code implementation21 May 2018 Feliks Hibraj, Sebastiano Vascon, Thilo Stadelmann, Marcello Pelillo

We report on a comprehensive set of experiments on the TIMIT dataset against standard clustering techniques and specific speaker clustering methods.

Sound Audio and Speech Processing

DeepScores -- A Dataset for Segmentation, Detection and Classification of Tiny Objects

2 code implementations27 Mar 2018 Lukas Tuggener, Ismail Elezi, Jürgen Schmidhuber, Marcello Pelillo, Thilo Stadelmann

We present the DeepScores dataset with the goal of advancing the state-of-the-art in small objects recognition, and by placing the question of object recognition in the context of scene understanding.

General Classification Object +3

Dominant Sets for "Constrained" Image Segmentation

no code implementations15 Jul 2017 Eyasu Zemene, Leulseged Tesfaye Alemu, Marcello Pelillo

In particular, we shall focus on interactive segmentation and co-segmentation (in both the unsupervised and the interactive versions).

Image Segmentation Interactive Segmentation +2

Multi-Target Tracking in Multiple Non-Overlapping Cameras using Constrained Dominant Sets

no code implementations19 Jun 2017 Yonatan Tariku Tesfaye, Eyasu Zemene, Andrea Prati, Marcello Pelillo, Mubarak Shah

In this paper, a unified three-layer hierarchical approach for solving tracking problems in multiple non-overlapping cameras is proposed.


On the Interplay between Strong Regularity and Graph Densification

no code implementations21 Mar 2017 Marco Fiorucci, Alessandro Torcinovich, Manuel Curado, Francisco Escolano, Marcello Pelillo

In this paper we analyze the practical implications of Szemer\'edi's regularity lemma in the preservation of metric information contained in large graphs.


Revealing Structure in Large Graphs: Szemerédi's Regularity Lemma and its Use in Pattern Recognition

no code implementations21 Sep 2016 Marcello Pelillo, Ismail Elezi, Marco Fiorucci

Introduced in the mid-1970's as an intermediate step in proving a long-standing conjecture on arithmetic progressions, Szemer\'edi's regularity lemma has emerged over time as a fundamental tool in different branches of graph theory, combinatorics and theoretical computer science.


Context Aware Nonnegative Matrix Factorization Clustering

no code implementations15 Sep 2016 Rocco Tripodi, Sebastiano Vascon, Marcello Pelillo

These interactions allow the players to choose a cluster which is coherent with the clusters chosen by similar players, a property which is not guaranteed by NMF, since it produces a soft clustering of the data.


Randomized Prediction Games for Adversarial Machine Learning

no code implementations3 Sep 2016 Samuel Rota Bulò, Battista Biggio, Ignazio Pillai, Marcello Pelillo, Fabio Roli

In spam and malware detection, attackers exploit randomization to obfuscate malicious data and increase their chances of evading detection at test time; e. g., malware code is typically obfuscated using random strings or byte sequences to hide known exploits.

BIG-bench Machine Learning General Classification +3

Interactive Image Segmentation Using Constrained Dominant Sets

no code implementations1 Aug 2016 Eyasu Zemene, Marcello Pelillo

We propose a new approach to interactive image segmentation based on some properties of a family of quadratic optimization problems related to dominant sets, a well-known graph-theoretic notion of a cluster which generalizes the concept of a maximal clique to edge-weighted graphs.

Image Segmentation Semantic Segmentation

Document Clustering Games in Static and Dynamic Scenarios

no code implementations8 Jul 2016 Rocco Tripodi, Marcello Pelillo

Each document to be clustered is represented as a player and each cluster as a strategy.


A Game-Theoretic Approach to Word Sense Disambiguation

no code implementations CL 2017 Rocco Tripodi, Marcello Pelillo

This paper presents a new model for word sense disambiguation formulated in terms of evolutionary game theory, where each word to be disambiguated is represented as a node on a graph whose edges represent word relations and senses are represented as classes.

Semantic Similarity Semantic Textual Similarity +1

Context-Sensitive Decision Forests for Object Detection

no code implementations NeurIPS 2012 Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof

In this paper we introduce Context-Sensitive Decision Forests - A new perspective to exploit contextual information in the popular decision forest framework for the object detection problem.

General Classification Object +4

A Game-Theoretic Approach to Hypergraph Clustering

no code implementations NeurIPS 2009 Samuel R. Bulò, Marcello Pelillo

In contrast to the classical approach, we attempt to provide a meaningful formalization of the very notion of a cluster and we show that game theory offers an attractive and unexplored perspective that serves well our purpose.


Cannot find the paper you are looking for? You can Submit a new open access paper.