Publications associées à Apprentissage organisationnel

On the (In)security of Peer-to-Peer Decentralized Machine Learning

Carmela González Troncoso, Mathilde Aliénor Raynal, Dario Pasquini

In this work, we carry out the first, in-depth, privacy analysis of Decentralized Learning-a collaborative machine learning framework aimed at addressing the main limitations of federated learning. We introduce a suite of novel attacks for both passive and ...

IEEE COMPUTER SOC2023

Improving Generalization of Pretrained Language Models

Rabeeh Karimi Mahabadi

In this dissertation, we propose multiple methods to improve transfer learning for pretrained language models (PLMs). Broadly, transfer learning is a powerful technique in natural language processing, where a language model is first pre-trained on a data-r ...

EPFL2023

Beyond Spectral Gap: The Role of the Topology in Decentralized Learning

Martin Jaggi, Thijs Vogels, Hadrien Hendrikx

In data-parallel optimization of machine learning models, workers collaborate to improve their estimates of the model: more accurate gradients allow them to use larger learning rates and optimize faster. In the decentralized setting, in which workers commu ...

Microtome Publishing2023

Lifelong Machine Learning with Data Efficiency and Knowledge Retention

Fei Mi

Artificial intelligence (AI) and machine learning (ML) have become de facto tools in many real-life applications to offer a wide range of benefits for individuals and our society. A classic ML model is typically trained with a large-scale static dataset in ...

EPFL2021

Word-level Embeddings for Cross-Task Transfer Learning in Speech Processing

Pierre Anton Beckmann, Milos Cernak

Recent breakthroughs in deep learning often rely on representation learning and knowledge transfer. In recent years, unsupervised and self-supervised techniques for learning speech representation were developed to foster automatic speech recognition. Up to ...

EUROPEAN ASSOC SIGNAL SPEECH & IMAGE PROCESSING-EURASIP2021

Critical Parameters for Scalable Distributed Learning with Large Batches and Asynchronous Updates

Martin Jaggi, Amirkeivan Mohtashami, Sebastian Urban Stich

It has been experimentally observed that the efficiency of distributed training with stochastic gradient (SGD) depends decisively on the batch size and—in asynchronous implementations—on the gradient staleness. Especially, it has been observed that the spe ...

MICROTOME PUBLISHING2021

Approaching Ontology Alignment through Representation Learning to Bridge the Semantic Gap in Engineering Applications

Prodromos Kolyvakis

The current information landscape is characterised by a vast amount of relatively semantically homogeneous, when observed in isolation, data silos that are, however, drastically semantically fragmented when considered as a whole. Within each data silo, inf ...

EPFL2020

Imagining with and in Space: Structural support as a cognitive device in architectural education

Dieter Dietz, Julien Lafontaine Carboni, Aurélie Dupuis, Darío Negueruela Del Castillo

In this paper, we will argue that the structure of the spatio-temporal pedagogical set-up and the elaboration of an evolutionary support for operations of cognition and imagination are powerful in transforming learning cultures. We develop this argument in ...

2020

Age-Dependent Statistical Learning Trajectories Reveal Differences in Information Weighting

Steffen Alexander Herff

Statistical learning (SL) is the ability to generate predictions based on probabilistic dependencies in the environment, an ability that is present throughout life. The effect of aging on SL is still unclear. Here, we explore statistical learning in health ...

AMER PSYCHOLOGICAL ASSOC2020

Knowledge Transfer with Jacobian Matching

François Fleuret, Suraj Srinivas

Classical distillation methods transfer representations from a “teacher” neural network to a “student” network by matching their output activations. Recent methods also match the Jacobians, or the gradient of output activations with the input. However, thi ...

2018

Monitoring Collaborative Learning Activities: Exploring the Differential Value of Collaborative Flow Patterns for Learning Analytics

Maria Jesus Rodriguez Triana

Collaborative learning flow patterns (CLFPs) encode solutions to recurrent pedagogical problems, which have been successfully applied to the design of learning experiences. However, the pedagogical knowledge encoded in these patterns has seldom been exploi ...

2018

Semantically Enriched Industry Data & Information Modelling: A feasibility study on Shop-floor Incident Recognition

Dimitrios Kyritsis, Damiano Nunzio Arena, Apostolos Perdikakis

Knowledge modelling at industrial level consists an importunate activity nowadays due to the ceaseless advances in technologies and standards applied as well as the extensive amount of unrelated real-time and historical data at shop-floor level. A Common I ...

IEEE2016

Review of Advances in Neural Networks: Neural Design Technology Stack

Yusuf Leblebici, Stanislaw Andrzej Wozniak

This review provides a high-level synthesis of significant recent advances in artificial neural network research, as well as multi-disciplinary concepts connected to the far-reaching goal of obtaining intelligent systems. We assume that a global outlook of ...

Elsevier2016

Exploring the effects of intellectual capital architectures on organizational ambidexterity. An fsQCA approach

Lorenzo Massa

Purpose - As the global market conditions become progressively more volatile and unpredictable, public and private organisations are increasingly pressured to rely on innovation and adaptability as crucial sources of sustained performance. Our study invest ...

Ikam-Inst Knowledge Asset Management2015

An integrated way of using a tangible user interface in a classroom

Pierre Dillenbourg, Sébastien Cuendet, Jessica Elke Dehler Zufferey

Despite many years of research in CSCL, computers are still scarcely used in classrooms today. One reason for this is that the constraints of the classroom environment are neglected by designers. In this contribution, we present a CSCL environment designed ...

2015

Knowledge Recombination Across Technological Boundaries: Scientists vs. Engineers

Marc Gruber

Building on the seminal work of Thomas J. Allen, we contribute to the emerging microlevel theory of knowledge recombination by examining how individual-level characteristics of inventors affect the breadth of their technological recombinations. Our data se ...

INFORMS2013

Learning Analytics Dashboard Applications

Sten Govaerts

This article introduces learning analytics dashboards that visualize learning traces for learners and teachers. We present a conceptual framework that helps to analyze learning analytics applications for these kinds of users. We then present our own work i ...

Sage Publications Inc2013

Escaping the Prior Knowledge Corridor: What shapes the Number and Variety of Market Opportunities identified before Market Entry of Technology Start-ups?

Marc Gruber, James David Thompson

The choice of the firm's market environment is one of the fundamental decisions of firm founders. We study the pre-entry generation of founders' market choice sets by investigating their search for market opportunities in which the firm's technological res ...

INFORMS2013

Putting the user into the active learning loop: towards realistic but efficient photointerpretation

Devis Tuia

In recent years, several studies have been published about the smart definition of training set using active learning algorithms. However, none of these works consider the contradiction between the active learning methods, which rank the pixels according t ...

Ieee2012

Tackling Engineering Education Research Challenges: Web 2.0 Social Software for Personal Learning

Denis Gillet

This paper focuses on Engineering Education Research on Technology Enhanced Learning carried out at the Swiss Federal Institute of Technology in Lausanne (EPFL), and on its current focus on personal and collaborative learning. After some thoughts on the di ...

Tempus Publications2010