Skip to content
ICML paper 'High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approach'
Branch: master
Clone or download
Latest commit c9996df May 15, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
code Add files via upload Jun 12, 2018
images Add files via upload Feb 21, 2018
QD_AsFastAsPoss_notebook.ipynb Add files via upload Jun 19, 2018
README.md Update README.md May 15, 2019

README.md

High-Quality Prediction Intervals for Deep Learning

Code accompanying the ICML 2018 paper High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approach https://arxiv.org/abs/1802.07167.

Intro

How can we get uncertainty estimates from deep learning systems?

Estimating model uncertainty.

Comparison against MVE.

Jupyter Notebook Keras Demo

A simple fast demo using Keras is included in QD_AsFastAsPoss_notebook.ipynb.

Code Structure

Main paper code in 5 files:

  • main.py
  • pso.py
  • DataGen.py
  • DeepNetPI.py
  • utils.py
  • inputs.txt

Run main.py to reproduce first figure.

We have included hyperparameters used for the boston and concrete datasets in inputs.txt.

You can’t perform that action at this time.