Publications associées à Biais (distorsion)

Exploiting the Signal-Leak Bias in Diffusion Models

Sabine Süsstrunk, Radhakrishna Achanta, Mahmut Sami Arpa, Martin Nicolas Everaert, Athanasios Fitsios

There is a bias in the inference pipeline of most diffusion models. This bias arises from a signal leak whose distribution deviates from the noise distribution, creating a discrepancy between training and inference processes. We demonstrate that this signa ...

2024

It’s All Relative: Learning Interpretable Models for Scoring Subjective Bias in Documents from Pairwise Comparisons

Matthias Grossglauser, Aswin Suresh, Chi Hsuan Wu

We propose an interpretable model to score the subjective bias present in documents, based only on their textual content. Our model is trained on pairs of revisions of the same Wikipedia article, where one version is more biased than the other. Although pr ...

2024

Biases in Information Selection and Processing: Survey Evidence from the Pandemic

Andreas Fuster

We conduct two survey experiments to study which information people choose to consume and how it affects their beliefs. In the first experiment, respondents choose between optimistic and pessimistic article headlines related to the COVID-19 pandemic and ar ...

2024

Serial Dependence in Human Visual Perception and Decision-Making

Gizay Ceylan

Decisions about a current visual stimulus are systematically biased by recently encountered stimuli, a phenomenon known as serial dependence. In human vision, for instance, we tend to report the features of current images as more similar â i.e., an attra ...

EPFL2024

Hybrid Simulator for Capturing Dynamics of Synthetic Populations

Michel Bierlaire, Marija Kukic

This paper presents a novel hybrid framework for generating and updating a synthetic population. We call it hybrid because it combines model-based and data-driven approaches. Existing generators produce a snapshot of synthetic data that becomes outdated ov ...

IEEE2024

A Research-Practice Partnership to Introduce Computer Science in Secondary School: Lessons from a Pilot Program

Francesco Mondada, Helena Kovacs, Jean-Philippe Pellet, Jessica Elke Dehler Zufferey, Barbara Bruno, Laila Abdelsalam El-Hamamsy

Context: Introducing Computer Science (CS) into formal education can be challenging, notably when considering the numerous stakeholders involved which include the students, teachers, schools, and policy makers. We believe these perspectives should be con ...

2023

How Words Move Hearts: Interpretable Machine Learning Models of Bias, Engagement, and Influence in Socio-Political Systems

Aswin Suresh

We study socio-political systems in representative democracies. Motivated by problems that affect the proper functioning of the system, we build computational methods to answer research questions regarding the phenomena occurring in them. For each phenomen ...

EPFL2023

Timing and spatial selection bias in rapid extreme event attribution

Anthony Christopher Davison, Ophélia Mireille Anna Miralles

Selection bias may arise when data have been chosen in a way that subsequent analysis does not account for. Such bias can arise in climate event attribution studies that are performed rapidly after a devastating "trigger event'', whose occurrence correspon ...

ELSEVIER2023

The inductive bias of deep learning: Connecting weights and functions

Guillermo Ortiz Jimenez

Years of a fierce competition have naturally selected the fittest deep learning algorithms. Yet, although these models work well in practice, we still lack a proper characterization of why they do so. This poses serious questions about the robustness, trus ...

EPFL2023

Stronger serial dependence in the depth plane than the fronto-parallel plane between realistic objects: Evidence from virtual reality

David Pascucci

Visual estimates of stimulus features are systematically biased toward the features of previously encountered stimuli. Such serial dependencies have often been linked to how the brain maintains perceptual continuity. However, serial dependence has mostly b ...

2023

Incremental Learning in Diagonal Linear Networks

Raphaël Jean Berthier

Diagonal linear networks (DLNs) are a toy simplification of artificial neural networks; they consist in a quadratic reparametrization of linear regression inducing a sparse implicit regularization. In this paper, we describe the trajectory of the gradient ...

Microtome Publishing2023

The Impact of Data Persistence Bias on Social Media Studies

Tugrulcan Elmas

Social media studies often collect data retrospectively to analyze public opinion. Social media data may decay over time and such decay may prevent the collection of the complete dataset. As a result, the collected dataset may differ from the complete data ...

Assoc Computing Machinery2023

Calibration Transfer Methodology for Cloud Radars Based on Ice Cloud Observations

Alexis Berne, Anne-Claire Marie Billault--Roux

This article presents a calibration transfer methodology that can be used between radars of the same or dif-ferent frequency bands. This method enables the absolute calibration of a cloud radar by transferring it from another collocated instrument with kno ...

AMER METEOROLOGICAL SOC2023

Information Processing Biases: The Effects of Negative Emotional Symptoms on Sampling Pleasant and Unpleasant Information

Steffen Alexander Herff

Although theories of emotion associate negative emotional symptoms with cognitive biases in information processing, they rarely specify the details. Here, we characterize cognitive biases in information processing of pleasant and unpleasant information, an ...

AMER PSYCHOLOGICAL ASSOC2022

Bias at a Second Glance: A Deep Dive into Bias for German Educational Peer-Review Data Modeling

Vinitra Swamy, Tanja Christina Käser Jacober, Thiemo Wambsganss

Natural Language Processing (NLP) has become increasingly utilized to provide adaptivity in educational applications. However, recent research has highlighted a variety of biases in pre-trained language models. While existing studies investigate bias in di ...

2022

Perturbation theory models for LSST-era galaxy clustering: Tests with subpercent mock catalog measurements in Fourier and configuration space

Jonathan Andrew Blazek

We analyze the clustering of galaxies using the z = 1.006 snapshot of the CosmoDC2 simulation, a high-fidelity synthetic galaxy catalog designed to validate analysis methods for the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST). We prese ...

American Physical Society2022

Learning Disentangled Behaviour Patterns for Wearable-based Human Activity Recognition

Tao Lin

In wearable-based human activity recognition (HAR) research, one of the major challenges is the large intra-class variability problem. The collected activity signal is often, if not always, coupled with noises or bias caused by personal, environmental, or ...

ASSOC COMPUTING MACHINERY2022

Serab: A Multi-Lingual Benchmark For Speech Emotion Recognition

Pierre Anton Beckmann, Milos Cernak

Recent developments in speech emotion recognition (SER) often leverage deep neural networks (DNNs). Comparing and benchmarking different DNN models can often be tedious due to the use of different datasets and evaluation protocols. To facilitate the proces ...

IEEE2022

Towards Understanding Sharpness-Aware Minimization

Nicolas Henri Bernard Flammarion, Maksym Andriushchenko

Sharpness-Aware Minimization (SAM) is a recent training method that relies on worst-case weight perturbations which significantly improves generalization in various settings. We argue that the existing justifications for the success of SAM which are based ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2022

Dynamics of the most common pathogenic mtDNA variant m.3243A > G demonstrate frequency-dependency in blood and positive selection in the germline

Konstantin Popadin

The A-to-G point mutation at position 3243 in the human mitochondrial genome (m.3243A > G) is the most common pathogenic mtDNA variant responsible for disease in humans. It is widely accepted that m.3243A > G levels decrease in blood with age, and an age c ...

Oxford University Press2022