miscellaneous-7

Uncertainty Quantification

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in uncertainty-quantification-7

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find uncertainty-quantification-7 models and implementations

google/uncertainty-baselines

4 papers 1,360

Datasets

No datasets available.

Subtasks

No subtasks available.

Most implemented papers

Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules

Johannes Klicpera, Stephan Gunnemann, Shankari Giri, Johannes T. Margraf•Fri Nov 27 2020

The DimeNet++ model is proposed, which is 8x faster and 10% more accurate than the original Dime net on the QM9 benchmark of equilibrium molecules, and ensembling and mean-variance estimation for uncertainty quantification are investigated with the goal of accelerating the exploration of the vast space of non-equilibrium structures.

387

Content

lightning-uq-box/lightning-uq-box

4 papers 72

yihaosun1124/OfflineRL-Kit

2 papers 227

faceonlive/ai-research

2 papers 139

slimgroup/Software.SEG2020

2 papers 15

0

Paper Graph

Uncertainty Sets for Image Classifiers using Conformal Prediction

Michael I. Jordan, J. Malik, Anastasios Nikolas Angelopoulos, Stephen Bates•Mon Sep 28 2020

An algorithm is presented that modifies any classifier to output a predictive set containing the true label with a user-specified probability, such as 90%, which provides a formal finite-sample coverage guarantee for every model and dataset.

427 0

Paper Graph

Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness

Balaji Lakshminarayanan, Dustin Tran, Zi Lin, Shreyas Padhy, Jeremiah Zhe Liu, Tania Bedrax-Weiss•Sun May 31 2020

Spectral-normalized Neural Gaussian Process (SNGP), a simple method that improves the distance-awareness ability of modern DNNs, by adding a weight normalization step during training and replacing the output layer with a Gaussian process and outperforms the other single-model approaches.

525 0

Paper Graph

Rule-based Bayesian regression

T. Botsas, L. Mason, Indranil Pan•Sat Aug 01 2020

A novel rule-based approach for handling regression problems that carries elements from two frameworks: it provides information about the uncertainty of the parameters of interest using Bayesian inference, and it allows the incorporation of expert knowledge through rule- based systems.

4 0

Paper Graph

COMBO: Conservative Offline Model-Based Policy Optimization

Chelsea Finn, S. Levine, Aviral Kumar, A. Rajeswaran, Tianhe Yu, Rafael Rafailov•Mon Feb 15 2021

This work develops a new model-based offline RL algorithm, COMBO, that regularizes the value function on out-of-support state-action tuples generated via rollouts under the learned model, and finds that it consistently performs as well or better as compared to prior offline model-free and model- based methods on widely studied offline RL benchmarks, including image-based tasks.

487 0

Paper Graph

A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification

Anastasios Nikolas Angelopoulos, Stephen Bates•Wed Jul 14 2021

This hands-on introduction is aimed to provide the reader a working understanding of conformal prediction and related distribution-free uncertainty quantification techniques with one self-contained document.

866 0

Paper Graph

Deep Uncertainty Quantification: A Machine Learning Approach for Weather Forecasting

Tianrui Li, Huaishao Luo, Jie Lu, Zheng Yan, Yu Zheng, Guangquan Zhang•Fri Dec 21 2018

This paper designs a data-driven method augmented by an effective information fusion mechanism to learn from historical data that incorporates prior knowledge from NWP by proposing a novel negative log-likelihood error (NLE) loss function.

160 0

Paper Graph

Deep learning observables in computational fluid dynamics

K. Lye, Siddhartha Mishra, Deep Ray•Thu Mar 07 2019

Under the assumption that the underlying neural networks generalize well, it is proved that the deep learning MC and QMC algorithms are guaranteed to be faster than the baseline (quasi-) Monte Carlo methods.

170 0

Paper Graph

Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels

A. Storkey, Massimiliano Patacchiola, Jack Turner, Elliot J. Crowley, M. O’Boyle•Mon Oct 12 2020

This work empirically demonstrate that DKT outperforms several state-of-the-art algorithms in few-shot classification, and is the state of the art for cross-domain adaptation and regression.

153 0

Paper Graph

Adding a benchmark result helps the community track progress.