DETECTION OF ADVERSARIAL ATTACKS USING DEEP LEARNING AND FEATURES EXTRACTED FROM INTERPRETABILITY METHODS IN INDUSTRIAL SCENARIOS

Detection of Adversarial Attacks Using Deep Learning and Features Extracted From Interpretability Methods in Industrial Scenarios

The adversarial training technique has been shown to improve the robustness of Machine Learning and Deep Learning models to adversarial attacks in the Computer Vision field.However, the effectiveness of this approach needs to be proven in the field of Anomaly Detection on industrial environments, where adversarial training has critical limitations.

read more


Challenges for Children with Cochlear Implants in Everyday Listening Scenarios: The Competitive Effect of Noise and Face Masks on Speech Intelligibility

Speech intelligibility (SI) tests under realistic acoustic scenarios are complex tasks to perform.Optimal acoustics, in terms of reverberation and noise, are thus needed.This is particularly true in the presence of young hard-of-hearing (HoH) children equipped with cochlear implants who need speech to be highly intelligible to learn.During the COVI

read more

Assessment of Treatment Plan Quality between Flattening Filter and Flattening Filter Free Photon Beam for Carcinoma of the Esophagus with IMRT Technique

Background: As compared to the flattened photon beam, removing the flattening filter (FF) from the head of a gantry decreases the average energy of the photon beam and increases the dose rate, leading to an impact on the quality of treatment plans.Objective: This study aimed to compare the quality of intensity-modulated radiation therapy (IMRT) tre

read more