Publication Library
The Security of Deep Learning Defences for Medical Imaging
Description: Deep learning has shown great promise in the domain of medical image analysis. Medical professionals and healthcare providers have been adopting the technology to speed up and enhance their work. These systems use deep neural networks (DNN) which are vulnerable to adversarial samples; images with imperceivable changes that can alter the model’s prediction. Researchers have proposed defences which either make a DNN more robust or detect the adversarial samples before they do harm. However, none of these works consider an informed attacker which can adapt to the defence mechanism. We show that an informed attacker can evade five of the current state of the art defences while successfully fooling the victim’s deep learning model, rendering these defences useless. We then suggest better alternatives for securing healthcare DNNs from such attacks: (1) harden the system’s security and (2) use digital signatures.
Created At: 14 December 2024
Updated At: 14 December 2024
Data Representativity for Machine Learning and AI Systems
Description: Data Representativity for Machine Learning and AI Systems
Created At: 14 December 2024
Updated At: 14 December 2024
Reinforcement Learning for Precision Oncology
Description: Reinforcement Learning for Precision Oncology
Created At: 14 December 2024
Updated At: 14 December 2024
Seven Tools of Causal Inference with Reflections on Machine Learning
Description: Seven Tools of Causal Inference with Reflections on Machine Learning
Created At: 14 December 2024
Updated At: 14 December 2024
Evolutionary Finance Simulations and the Agent-Based Approach
Description: The only way to make a major advance in finance modeling is to explore entirely new approaches rather than make incremental modifications to existing models. The purpose of our research is precisely to build models and perform analysis of the economy as a complex system prone to sudden and major changes. This includes the collection of new methods of empirical analysis, and the development of new mathematical and computational tools. This effort will be guided by emerging new conceptual paradigms such as network theory, behavioral economics and agent-based modeling, and empirically grounded by laboratory experiments and micro data. By combining these ideas into a practical simulation and forecasting tool, we aim at building the foundations of a new paradigm within finance.
Created At: 14 December 2024
Updated At: 14 December 2024