Publications related to AI alignment

Voices from the Frontline: Revealing the AI Practitioners' viewpoint on the European AI Act

Artificial intelligence (AI) is increasingly used in an ever larger number of industries. Alongside this development, however, abundant works argue that AI-driven systems are lacking in terms of safety, ethics and transparency. As a direct consequence, the ...

2024

When Your AI Becomes a Target: AI Security Incidents and Best Practices

Alexandre Massoud Alahi, Kathrin Grosse

In contrast to vast academic efforts to study AI security, few real-world reports of AI security incidents exist. Released incidents prevent a thorough investigation of the attackers' motives, as crucial information about the company and AI application is ...

AAAI Press2024

EdgeAI-Aware Design of In-Memory Computing Architectures

Marco Antonio Rios

Driven by the demand for real-time processing and the need to minimize latency in AI algorithms, edge computing has experienced remarkable progress. Decision-making AI applications stand out for their heavy reliance on data-centric operations, predominantl ...

EPFL2024

Comparing various AI approaches to traditional quantitative assessment of the myocardial perfusion in [82Rb] PET for MACE prediction

Julien René Pierre Fageot, Adrien Raphaël Depeursinge, Daniel Abler

Assessing the individual risk of Major Adverse Cardiac Events (MACE) is of major importance as cardiovascular diseases remain the leading cause of death worldwide. Quantitative Myocardial Perfusion Imaging (MPI) parameters such as stress Myocardial Blood F ...

Nature Portfolio2024

Exploring High-Performance and Energy-Efficient Architectures for Edge AI-Enabled Applications

Joshua Alexander Harrison Klein

The desire and ability to place AI-enabled applications on the edge has grown significantly in recent years. However, the compute-, area-, and power-constrained nature of edge devices are stressed by the needs of the AI-enabled applications, due to a gener ...

EPFL2024

Language Model Decoding as Likelihood–Utility Alignment

Boi Faltings, Robert West, Maxime Jean Julien Peyrard, Martin Josifoski, Valentin Hartmann, Debjit Paul, Jiheng Wei, Frano Rajic

A critical component of a successful language generation pipeline is the decoding algorithm. However, the general principles that should guide the choice of a decoding algorithm re- main unclear. Previous works only compare decoding algorithms in narrow sc ...

Assoc Computational Linguistics-Acl2023

Towards more Practical Threat Models in Artificial Intelligence Security

Alexandre Massoud Alahi, Kathrin Grosse

Recent works have identified a gap between research and practice in artificial intelligence security: threats studied in academia do not always reflect the practical use and security risks of AI. For example, while models are often studied in isolation, th ...

Usenix Association2024

Out of their minds? Externalist challenges for using AI in forensic psychiatry

Marcello Ienca, Ambra D'imperio

Harnessing the power of machine learning (ML) and other Artificial Intelligence (AI) techniques promises substantial improvements across forensic psychiatry, supposedly offering more objective evaluations and predictions. However, AI-based predictions abou ...

FRONTIERS MEDIA SA2023

Robust Training and Verification of Deep Neural Networks

Fabian Ricardo Latorre Gomez

According to the proposed Artificial Intelligence Act by the European Comission (expected to pass at the end of 2023), the class of High-Risk AI Systems (Title III) comprises several important applications of Deep Learning like autonomous driving vehicles ...

EPFL2023

Overflow-free compute memories for edge AI acceleration

David Atienza Alonso, Giovanni Ansaloni, Alexandre Sébastien Julien Levisse, Marco Antonio Rios, Flavio Ponzina

Compute memories are memory arrays augmented with dedicated logic to support arithmetic. They support the efficient execution of data-centric computing patterns, such as those characterizing Artificial Intelligence (AI) algorithms. These architectures can ...

2023

Bit-Line Computing for CNN Accelerators Co-Design in Edge AI Inference

David Atienza Alonso, Giovanni Ansaloni, Alexandre Sébastien Julien Levisse, Marco Antonio Rios, Flavio Ponzina

By supporting the access of multiple memory words at the same time, Bit-line Computing (BC) architectures allow the parallel execution of bit-wise operations in-memory. At the array periphery, arithmetic operations are then derived with little additional o ...

2023

Artificial Neural Network Training on an Optical Processor via Direct Feedback Alignment

Florent Gérard Krzakala, Julien Marcel Daniel Emmanuel Launay

Artificial Neural Networks (ANN) are habitually trained via the back-propagation (BP) algorithm. This approach has been extremely successful: Current models like GPT-3 have O(10 11 ) parameters, are trained on O(10 11 ) words and produce awe-inspiring resu ...

IEEE2023

From event-based surprise to lifelong learning. A journey in the timescales of adaptation

Martin Louis Lucien Rémy Barry

Humans and animals constantly adapt to their environment over the course of their life. This thesis seeks to integrate various timescales of adaptation, ranging from the adaptation of synaptic connections between spiking neurons (milliseconds), rapid behav ...

EPFL2023

Augmented Lagrangian Methods for Provable and Scalable Machine Learning

Mehmet Fatih Sahin

Non-convex constrained optimization problems have become a powerful framework for modeling a wide range of machine learning problems, with applications in k-means clustering, large- scale semidefinite programs (SDPs), and various other tasks. As the perfor ...

EPFL2023

Exploiting Explanations to Detect Misclassifications of Deep Learning Models in Power Grid Visual Inspection

Olga Fink, Giovanni Floreale

In the context of automatic visual inspection of infrastructures by drones, Deep Learning (DL) models are used to automatically process images for fault diagnostics. While explainable Artificial Intelligence (AI) algorithms can provide explanations to asse ...

Research Publishing2023

Safe multi-agent deep reinforcement learning for joint bidding and maintenance scheduling of generation units

Olga Fink, Mina Montazeri

This paper proposes a safe reinforcement learning algorithm for generation bidding decisions and unit maintenance scheduling in a competitive electricity market environment. In this problem, each unit aims to find a bidding strategy that maximizes its reve ...

2023

Intentional machines: A defence of trust in medical artificial intelligence

Trust constitutes a fundamental strategy to deal with risks and uncertainty in complex societies. In line with the vast literature stressing the importance of trust in doctor–patient relationships, trust is therefore regularly suggested as a way of dealing ...

2022

Misplaced Trust and Distrust: How Not to Engage with Medical Artificial Intelligence

Marcello Ienca

Artificial intelligence (AI) plays a rapidly increasing role in clinical care. Many of these systems, for instance, deep learning-based applications using multilayered Artificial Neural Nets, exhibit epistemic opacity in the sense that they preclude compre ...

CAMBRIDGE UNIV PRESS2022

A 3D indicator for guiding AI applications in the energy sector

Christophe Ballif, Alessandro Francesco Aldo Virtuani, Nicolas Würsch, Hugo James André Quest, Fabian Heymann, Laure-Emmanuelle Perret Aebi, Christian Rod, Marine Dominique Cauz

The utilisation of Artificial Intelligence (AI) applications in the energy sector is gaining momentum, with increasingly intensive search for suitable, high-quality and trustworthy solutions that displayed promising results in research. The growing interes ...

2022

Nine recommendations for the governance of AI systems

Marie-Valentine Renée Agnès Florin

Some governance functions traditionally performed by humans are increasingly informed and sometimes automatically executed by machine learning algorithms (governance by machine learning) to benefit society. Therefore, it is necessary to think also about th ...

EPFL2022