Url: A representation learning benchmark for transferable uncertainty estimates

M Kirchhof, B Mucsányi, SJ Oh… - Advances in Neural …, 2023 - proceedings.neurips.cc
Abstract Representation learning has significantly driven the field to develop pretrained
models that can act as a valuable starting point when transferring to new datasets. With the …

Probabilistic contrastive learning recovers the correct aleatoric uncertainty of ambiguous inputs

M Kirchhof, E Kasneci, SJ Oh - International Conference on …, 2023 - proceedings.mlr.press
Contrastively trained encoders have recently been proven to invert the data-generating
process: they encode each input, eg, an image, into the true latent vector that generated the …

Multi-similarity contrastive learning

E Mu, J Guttag, M Makar - arxiv preprint arxiv:2307.02712, 2023 - arxiv.org
Given a similarity metric, contrastive methods learn a representation in which examples that
are similar are pushed together and examples that are dissimilar are pulled apart …

Enhancing Out-of-Distribution Detection Through Stochastic Embeddings in Self-supervised Learning

D Janiak, J Binkowski, P Bielak… - … on Computational Science, 2024 - Springer
In recent years, self-supervised learning has played a pivotal role in advancing machine
learning by allowing models to acquire meaningful representations from unlabeled data. An …

Unveiling the Potential of Probabilistic Embeddings in Self-Supervised Learning

D Janiak, J Binkowski, P Bielak… - arxiv preprint arxiv …, 2023 - arxiv.org
In recent years, self-supervised learning has played a pivotal role in advancing machine
learning by allowing models to acquire meaningful representations from unlabeled data. An …

Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstruction

S Wang, Y **e, CP Chang, C Millerdurai… - arxiv preprint arxiv …, 2024 - arxiv.org
Neural implicit fields have recently emerged as a powerful representation method for multi-
view surface reconstruction due to their simplicity and state-of-the-art performance …

Uncertainties of latent representations in computer vision

M Kirchhof - arxiv preprint arxiv:2408.14281, 2024 - arxiv.org
Uncertainty quantification is a key pillar of trustworthy machine learning. It enables safe
reactions under unsafe inputs, like predicting only when the machine learning model detects …

LiST: An All-Linear-Layer Spatial-Temporal Feature Extractor with Uncertainty Estimation for RUL Prediction

Z Huang, C Gruhl, B Sick - 2024 IEEE 19th Conference on …, 2024 - ieeexplore.ieee.org
In the context of Remaining Useful Life (RUL) prediction for industrial systems, the pursuit of
prediction accuracy must be balanced against the hardware costs of model operation and …

Quantifying Representation Reliability in Self-Supervised Learning Models

YJ Park, H Wang, S Ardeshir, N Azizan - arxiv preprint arxiv:2306.00206, 2023 - arxiv.org
Self-supervised learning models extract general-purpose representations from data.
Quantifying the reliability of these representations is crucial, as many downstream models …

LLM2Loss: Leveraging Language Models for Explainable Model Diagnostics

S Ardeshir - arxiv preprint arxiv:2305.03212, 2023 - arxiv.org
Trained on a vast amount of data, Large Language models (LLMs) have achieved
unprecedented success and generalization in modeling fairly complex textual inputs in the …