activity
20172020
most citedRobustness to Adversarial Examples through an Ensemble of Specialists

18 citations · 21 across the 3 of their papers we have counts for

collaborators

8 papers

cs.CV20203 cited

Self-supervised Robust Object Detectors from Partially Labelled Datasets

Mahdieh Abbasi, Denis Laurendeau, Christian Gagne

In the object detection task, merging various datasets from similar contexts but with different sets of Objects of Interest (OoI) is an inexpensive way (in terms of labor cost) for…

cs.LG2020

Toward Adversarial Robustness by Diversity in an Ensemble of Specialized Deep Neural Networks

Mahdieh Abbasi, Arezoo Rajabi, Christian Gagne +1

We aim at demonstrating the influence of diversity in the ensemble of CNNs on the detection of black-box adversarial instances and hardening the generation of white-box adversarial…

cs.LG2019

Toward Metrics for Differentiating Out-of-Distribution Sets

Mahdieh Abbasi, Changjian Shui, Arezoo Rajabi +2

Vanilla CNNs, as uncalibrated classifiers, suffer from classifying out-of-distribution (OOD) samples nearly as confidently as in-distribution samples. To tackle this challenge, som…

cs.LG2019

A Principled Approach for Learning Task Similarity in Multitask Learning

Changjian Shui, Mahdieh Abbasi, Louis-Émile Robitaille +2

Multitask learning aims at solving a set of related tasks simultaneously, by exploiting the shared knowledge for improving the performance on individual tasks. Hence, an important…

cs.CV2018

Controlling Over-generalization and its Effect on Adversarial Examples Generation and Detection

Mahdieh Abbasi, Arezoo Rajabi, Azadeh Sadat Mozafari +2

Convolutional Neural Networks (CNNs) significantly improve the state-of-the-art for many applications, especially in computer vision. However, CNNs still suffer from a tendency to…

cs.CR2018

Towards Dependable Deep Convolutional Neural Networks (CNNs) with Out-distribution Learning

Mahdieh Abbasi, Arezoo Rajabi, Christian Gagné +1

Detection and rejection of adversarial examples in security sensitive and safety-critical systems using deep CNNs is essential. In this paper, we propose an approach to augment CNN…