Publications | tango-horizon.eu

Enhancing Privacy and Utility in Federated Learning

Machine Learning (ML) methodologies require data to train models. Getting data is not always easy: users may not be prone to share their data with a third party Read More >>>

GLOR-FLEX: Local to Global Rule-based EXplanations for Federated Learning

The increasing spread of artificial intelligence applications has led to decentralized frameworks that foster collaborative model training among multiple entities. Read More >>>

Interpretable and Fair Mechanisms for Abstaining Classifiers

Abstaining classifiers have the option to refrain from providing a prediction for instances that are difficult to classify. The abstention mechanism is designed to trade off the classifier’s performance on the accepted data while ensuring a minimum number of predictions. Read More >>>

Can LLMs Correct Physicians, Yet?

Can LLMs Correct Physicians, Yet? Investigating Effective Interaction Methods in the Medical Domain

We explore the potential of Large Language Models (LLMs) to assist and potentially correct physicians in medical decision-making tasks. Read More >>>

SYNERGY 2024 – Designing and Building Hybrid Human-AI Systems 2024

Workshop on Designing and Building Hybrid Human-AI Systems, co-located with 17th International Conference on Advanced Visual Interfaces (AVI 2024). Read More >>>

Explaining the explainers in graph neural networks: A comparative study

Following a fast initial breakthrough in graph based learning, Graph Neural Networks (GNNs) have reached a widespread application in many science and engineering fields, prompting the need for methods to understand their decision process. Read More >>>

Addressing Attribute Bias with Adversarial Support-Matching

When trained on diverse labelled data, machine learning models have proven themselves to be a powerful tool in all facets of society. However, due to budget limitations, deliberate or non-deliberate censorship, and other problems Read More >>>

Coordination in Dynamic Interactions by Converging on Tacitly Agreed Joint Plans

People solve a myriad of coordination problems without explicit communication every day. A recent theoretical account, virtual bargaining, proposes that… Read More >>>

Conversational technology and reactions to withheld information

People frequently face decisions that require making inferences about withheld information. The advent of large language models coupled with conversational technology, e.g., Alexa, Siri, Cortana, and the Google Assistant, is changing the mode in which people make these inferences. Read More >>>

When rules are over-ruled

Virtual bargaining as a contractualist method of moral judgment

Rules help guide our behavior—particularly in complex social contexts. But rules sometimes give us the “wrong” answer. How do we know when it is okay to break the rules? Read More >>>

Uncertainty Matters: Uncertainty Matters: Stable Conclusions under Unstable Assessment of Fairness Results

Uncertainty Matters: Stable Conclusions under Unstable Assessment of Fairness Results

Recent studies highlight the effectiveness of Bayesian methods in assessing algorithm performance, particularly in fairness and bias evaluation. Read More >>>

A Frank System for Co-evolutionary Hybrid Decision Making

We introduce Frank, a human-in-the-loop system for co-evolutionary hybrid decision-making aiding the user to label records from an un-labeled dataset. Read More >>>

Amplified Contribution Analysis for Federated Learning

The problem of establishing the client’s marginal contribution is essential to any decentralised machine-learning process that relies on the participation of remote agents. Read More >>>

Designing for situated AI-human decision making

We present a case study of AI deployment in a UK primary care (family doctor) setting. This demonstrates some of the challenges of real-world deployment of AI-human systems for decision-making. Read More >>>

Epistemic Interaction – Tuning Interfaces to Provide Information for AI Support

As machine learning integrates deeper into human-computer interactions, the concept of epistemic interaction emerges, aiming to refine these interactions to enhance system adaptability. Read More >>>

Designing and Building Hybrid Human-AI Systems 2024

Proceedings of the 1st International Workshop on Designing and Building Hybrid Human-AI Systems, co-located with 17th International Conference on Advanced Visual Interfaces (AVI 2024) Read More >>>

Learning to Intervene on Concept Bottlenecks

While deep learning models often lack interpretability, concept bottleneck models (CBMs) provide inherent explanations via their concept representations. Read More >>>

Towards Logically Consistent Language Models

Towards Logically Consistent Language Models via Probabilistic Reasoning

Developing reliable large language models (LLMs) and safely deploying them is more and more crucial, particularly when they are used as external sources of knowledge… Read More >>>

Sheaf Diffusion Goes Nonlinear

Sheaf Diffusion Goes Nonlinear: Enhancing GNNs with Adaptive Sheaf Laplacians

Sheaf Neural Networks (SNNs) have recently been introduced to enhance Graph Neural Networks (GNNs) in their capability to learn from graphs Read More >>>

A Simple and Expressive Graph Neural Network Based Method for Structural Link Representation

Graph Neural Networks (GNNs) have achieved state-of-the-art results in tasks like node classification, link prediction, and graph classification. Read More >>>

The Evolution of Latent Representations in a Dynamic Knowledge Graph

Unveiling LLMs: The Evolution of Latent Representations in a Dynamic Knowledge Graph

Large Language Models (LLMs) demonstrate an impressive capacity to
recall a vast range of factual knowledge. Read More >>>

Preference Elicitation in Interactive and User-centered Algorithmic Recourse

Algorithmic Recourse aims to provide actionable explanations, or recourse plans, to overturn potentially unfavourable decisions taken by automated machine learning models Read More >>>

BEARS Make Neuro-Symbolic Models Aware of their Reasoning Shortcuts

Neuro-Symbolic (NeSy) predictors that conform to symbolic knowledge – encoding, e.g., safety constraints – can be affected by Reasoning Shortcuts. Read More >>>

Pix2Code: Learning to Compose Neural Visual Concepts as Programs

The challenge in learning abstract concepts from images in an unsupervised fashion lies in the required integration of visual perception and generalizable relational reasoning. Read More >>>

Personalized Algorithmic Recourse with Preference Elicitation

Algorithmic Recourse (AR) is the problem of computing a sequence of actions that– once performed Read More >>>

BEYOND THE FACE: BIOMETRICS AND SOCIETY

In January 2019, the Serbian Minister of Internal Affairs made a groundbreaking announcement on national television, revealing their collaboration with Huawei, the Chinese tech giant. Read More >>>

Resource-rational contractualism: A triple theory of moral cognition

It is widely agreed upon that morality guides people with conflicting interests towards agreements of mutual benefit. Read More >>>