Publications related to Cognitive architecture

Enabling Uncertainty Estimation in Iterative Neural Networks

Pascal Fua, Nikita Durasov, Doruk Oner, Minh Hieu Lê

Turning pass-through network architectures into iterative ones, which use their own output as input, is a well-known approach for boosting performance. In this paper, we argue that such architectures offer an additional benefit: The convergence rate of the ...

2024

Sparse autoregressive neural networks for classical spin systems

Giuseppe Carleo, Dian Wu, Indaco Biazzo

Efficient sampling and approximation of Boltzmann distributions involving large sets of binary variables, or spins, are pivotal in diverse scientific fields even beyond physics. Recent advances in generative neural networks have significantly impacted this ...

Iop Publishing Ltd2024

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Scott William Pesme

In this PhD manuscript, we explore optimisation phenomena which occur in complex neural networks through the lens of

2

-layer diagonal linear networks. This rudimentary architecture, which consists of a two layer feedforward linear network with a diagonal ...

EPFL2024

Towards practical reinforcement learning for tokamak magnetic control

Federico Alberto Alfredo Felici, Cristian Galperti, Jonas Buchli, Brendan Tracey

Reinforcement learning (RL) has shown promising results for real-time control systems, including the domain of plasma magnetic control. However, there are still significant drawbacks compared to traditional feedback control approaches for magnetic confinem ...

Lausanne2024

TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation

Jean-Philippe Thiran, Guillaume Marc Georges Vray, Devavrat Tomar

Most recent test-time adaptation methods focus on only classification tasks, use specialized network architectures, destroy model calibration or rely on lightweight information from the source domain. To tackle these issues, this paper proposes a novel Tes ...

IEEE2023

Inter-Task Transfer of Prism Adaptation through Motor Imagery

Lisa Aïcha Mireille Julie Fleury

Prism adaptation (PA) is a useful method to investigate short-term sensorimotor plasticity. Following active exposure to prisms, individuals show consistent after-effects, probing that they have adapted to the perturbation. Whether after-effects are transf ...

MDPI2023

Learning From Heterogeneous Data Based on Social Interactions Over Graphs

Ali H. Sayed, Stefan Vlaski, Virginia Bordignon

This work proposes a decentralized architecture, where individual agents aim at solving a classification problem while observing streaming features of different dimensions and arising from possibly different distributions. In the context of social learning ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2023

Classification of Tropical Deforestation Drivers with Machine Learning and Satellite Image Time Series

Devis Tuia, Jan Pisl, Marc Conrad Russwurm, Lloyd Haydn Hughes

Tropical deforestation is a major environmental problem with severe consequences such as carbon emissions or biodiversity loss. While much research focuses on monitoring and mapping deforestation, less attention is paid to understanding the various reasons ...

The Institute of Electrical and Electronics Engineers, Inc.2023

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

The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems

Indaco Biazzo

Autoregressive Neural Networks (ARNNs) have shown exceptional results in generation tasks across image, language, and scientific domains. Despite their success, ARNN architectures often operate as black boxes without a clear connection to underlying physic ...

Berlin2023

Cognitive architecture

Graph Chatbot

Chat with Graph Search

Enabling Uncertainty Estimation in Iterative Neural Networks

Sparse autoregressive neural networks for classical spin systems

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Towards practical reinforcement learning for tokamak magnetic control

TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation

Inter-Task Transfer of Prism Adaptation through Motor Imagery

Learning From Heterogeneous Data Based on Social Interactions Over Graphs

Classification of Tropical Deforestation Drivers with Machine Learning and Satellite Image Time Series

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

The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems

Sparse autoregressive neural networks for classical spin systems

Enabling Uncertainty Estimation in Iterative Neural Networks

Towards practical reinforcement learning for tokamak magnetic control

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Learning From Heterogeneous Data Based on Social Interactions Over Graphs

Inter-Task Transfer of Prism Adaptation through Motor Imagery

The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems

TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation

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

Classification of Tropical Deforestation Drivers with Machine Learning and Satellite Image Time Series