Computational statistics

Conformal prediction

Conformal prediction (CP) is a set of algorithms devised to assess the uncertainty of predictions produced by a machine learning model. CP algorithms do this by computing and comparing nonconformity measures (often referred to as α-values), of examples from the training set, and compare these with measure computed for examples from a test set. Conformal predictors can be divided into inductive and transductive. These mainly differ in their computational complexity and whether they can be applied to regression or classification tasks. Inductive algorithms train one or several machine learning models which are re-used for future test objects, and can be used for both classification and regression tasks, whereas transductive algorithms re-train the model for every test object, and can only be used for classification tasks. Conformal prediction requires a user-specified significance level for which the algorithm should produce its predictions. This significance level restricts the frequency of errors that the algorithm is allowed to make. For example, a significance level of 0.1 means that the algorithm can make at most 10% erroneous predictions. To meet this requirement, the output is a set prediction, instead of a point prediction produced by standard supervised machine learning models. For classification tasks, this means that predictions are not a single class, for example 'cat', but instead a set like {'cat', 'dog'}. Depending on how good is the underlying model (how well it can discern between cats, dogs and other animals) and the specified significance level, these sets can be smaller or larger. For regression tasks, the output is prediction intervals, where a smaller significance level (less allowed errors) produces wider intervals which are less specific, and vice versa – more allowed errors produces tighter prediction intervals. (Wikipedia).

Video thumbnail

Conformal Field Theory (CFT) | Infinitesimal Conformal Transformations

Conformal field theories are used in many areas of physics, from condensed matter physics, to statistical physics to string theory. They are defined as quantum field theories that are invariant under so-called conformal transformations. In this video, we will investigate these conformal tr

From playlist Particle Physics

Video thumbnail

Conformal Geometry Processing

Symposium on Geometry Processing 2017 Graduate School Lecture by Keenan Crane https://www.cs.cmu.edu/~kmcrane/ http://geometry.cs.ucl.ac.uk/SGP2017/?p=gradschool#abs_conformal_geometry Digital geometry processing is the natural extension of traditional signal processing to three-dimensi

From playlist Tutorials and Lectures

Video thumbnail

Conformal geometry processing

SGP2018 Graduate School | July 7-11 | Paris, France Speaker: Keenan Crane, Carnegie Mellon University Abstract: Digital geometry processing is the natural extension of traditional signal processing to three-dimensional geometric data. In recent years, methods based on so-called conformal

From playlist Tutorials and Lectures

Video thumbnail

Conformal Field Theory (CFT) | More on Infinitesimal Conformal Transformations

Conformal field theories are quantum field theories that are invariant under so-called conformal transformations. In this video, we will investigate these conformal transformations in three or more dimensions. More information and details can be found in the excellent book "Introduction

From playlist Particle Physics

Video thumbnail

How to Make Predictions in Regression Analysis

Please Subscribe here, thank you!!! https://goo.gl/JQ8Nys How to Make Predictions in Regression Analysis

From playlist Statistics

Video thumbnail

What are Conformal Mappings? | Nathan Dalaklis

Conformal Mappings are a gem of Complex analysis that play a big role in both the theory behind the analysis of functions of a complex variable as well as studying fluid dynamics and electrostatics in physics, along with general relativity. In this video, a brief introduction to these maps

From playlist The First CHALKboard

Video thumbnail

Rod Gover - An introduction to conformal geometry and tractor calculus (Part 4)

After recalling some features (and the value of) the invariant « Ricci calculus » of pseudo­‐Riemannian geometry, we look at conformal rescaling from an elementary perspective. The idea of conformal covariance is visited and some covariant/invariant equations from physics are recovered in

From playlist Ecole d'été 2014 - Analyse asymptotique en relativité générale

Video thumbnail

Assumption-free prediction intervals for black-box regression algorithms - Aaditya Ramdas

Seminar on Theoretical Machine Learning Topic: Assumption-free prediction intervals for black-box regression algorithms Speaker: Aaditya Ramdas Affiliation: Carnegie Mellon University Date: April 21, 2020 For more video please visit http://video.ias.edu

From playlist Mathematics

Video thumbnail

Rod Gover - An introduction to conformal geometry and tractor calculus (Part 1)

After recalling some features (and the value of) the invariant « Ricci calculus » of pseudo­‐Riemannian geometry, we look at conformal rescaling from an elementary perspective. The idea of conformal covariance is visited and some covariant/invariant equations from physics are recovered in

From playlist Ecole d'été 2014 - Analyse asymptotique en relativité générale

Video thumbnail

8 Fundamental Flavors and the sill of the Conformal Window by Anna Hasenfratz

PROGRAM NONPERTURBATIVE AND NUMERICAL APPROACHES TO QUANTUM GRAVITY, STRING THEORY AND HOLOGRAPHY (HYBRID) ORGANIZERS: David Berenstein (University of California, Santa Barbara, USA), Simon Catterall (Syracuse University, USA), Masanori Hanada (University of Surrey, UK), Anosh Joseph (II

From playlist NUMSTRING 2022

Video thumbnail

Recent progress in predictive inference - Emmanuel Candes, Stanford University

Emmanuel Candes - Stanford University Machine learning algorithms provide predictions with a self-reported confidence score, but they are frequently inaccurate and uncalibrated, limiting their use in sensitive applications. This talk introduces novel calibration techniques addressing two

From playlist Interpretability, safety, and security in AI

Video thumbnail

Verifying the predictions of Conformal Bootstrap through lattice calculations by Prasad Hegde

Bangalore Area Strings Meeting - 2017 TIME : 31 July 2017 to 02 August 2017 VENUE:Madhava Lecture Hall, ICTS Bangalore Bengaluru now has a large group of string theorists, with 9 faculty members in the area, between ICTS and IISc. This is apart from a large group of postdocs and graduate

From playlist Bangalore Area Strings Meeting - 2017

Video thumbnail

Rod Gover - An introduction to conformal geometry and tractor calculus (Part 3)

After recalling some features (and the value of) the invariant « Ricci calculus » of pseudo­‐Riemannian geometry, we look at conformal rescaling from an elementary perspective. The idea of conformal covariance is visited and some covariant/invariant equations from physics are recovered in

From playlist Ecole d'été 2014 - Analyse asymptotique en relativité générale

Video thumbnail

Alessandro Vichi - Anatomy of the Ising model from Conformal Bootstrap

The Ising model is the one the simplest and yet non-trivial Conformal Field Theory. For decades it has been a dream to study such an intricate strongly coupled theory non-perturbatively using symmetries and other consistency conditions. This idea, called the conformal bootstrap, saw some s

From playlist 100…(102!) Years of the Ising Model

Video thumbnail

Introduction to Lattice Field Theory (Lecture 3) by Anna Hasenfratz

PROGRAM NONPERTURBATIVE AND NUMERICAL APPROACHES TO QUANTUM GRAVITY, STRING THEORY AND HOLOGRAPHY (HYBRID) ORGANIZERS: David Berenstein (University of California, Santa Barbara, USA), Simon Catterall (Syracuse University, USA), Masanori Hanada (University of Surrey, UK), Anosh Joseph (II

From playlist NUMSTRING 2022

Video thumbnail

Inflation and Supergravity - A. Linde - 5/17/2014

Workshop on Primordial Gravitational Waves and Cosmology (May 16 - 17, 2014) Learn more about this workshop: http://burkeinstitute.caltech.edu/workshops Produced in association with Caltech Academic Media Technologies. © 2014 California Institute of Technology

From playlist Walter Burke Institute for Theoretical Physics - Workshop on Primordial Gravitational Waves and Cosmology (May 16 - 17, 2014)

Video thumbnail

Claire S. Adjiman (10/214/22): A hierarchical approach to crystal structure prediction

Title: A hierarchical approach to crystal structure prediction: status, applications and challenges Abstract: The prediction of the possible crystal structure(s) of organic molecules is an important scientific problem with numerous industrial applications, for instance in the pharmaceutic

From playlist AATRN/STMS

Video thumbnail

Rod Gover - An introduction to conformal geometry and tractor calculus (Part 2)

After recalling some features (and the value of) the invariant « Ricci calculus » of pseudo­‐Riemannian geometry, we look at conformal rescaling from an elementary perspective. The idea of conformal covariance is visited and some covariant/invariant equations from physics are recovered in

From playlist Ecole d'été 2014 - Analyse asymptotique en relativité générale

Related pages

Statistical hypothesis testing | Statistical classification | Regression analysis | Convolutional neural network | Cross-validation (statistics) | Softmax function | Inductive reasoning | Independent and identically distributed random variables