DOCSLIB.ORG

Non-Parametric Calibration for Classification

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Non-Parametric Calibration for Classification DEGREE PROJECT IN ENGINEERING PHYSICS, SECOND CYCLE, 30 CREDITS STOCKHOLM, SWEDEN 2019 Non-Parametric Calibration for Classification JONATHAN WENGER KTH ROYAL INSTITUTE OF TECHNOLOGY SCHOOL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE Kungliga Tekniska högskolan (KTH) Teknisk fysik Degree Project Non-Parametric Calibration for Classification Jonathan Wenger [email protected] Supervisor: Prof. Dr. Hedvig Kjellström Examiner: Prof. Dr. Danica Kragic Submission Date: July 3, 2019 Abstract Many applications for classification methods not only require high accuracy but also reliable estimation of predictive uncertainty. This is of particular importance in fields such as computer vision or robotics, where safety-critical decisions are made based on classification outcomes. However, while many current classification frameworks, in particular deep neural network architectures, provide very good results in terms of accuracy, they tend to incorrectly estimate their predictive uncertainty. In this thesis we focus on probability calibration, the notion that a classifier’s confi- dence in a prediction matches the empirical accuracy of that prediction. We study calibration from a theoretical perspective and connect it to over- and underconfi- dence, two concepts first introduced in the context of active learning. The main contribution of this work is a novel algorithm for classifier calibration. We propose a non-parametric calibration method which is, in contrast to existing approaches, based on a latent Gaussian process and specifically designed for multi- class classification. It allows for the incorporation of prior knowledge, can be applied to any classification method that outputs confidence estimates and is not limited to neural networks. We demonstrate the universally strong performance of our method across different classifiers and benchmark data sets from computer vision in comparison to exist- ing classifier calibration techniques. Finally, we empirically evaluate the effects of calibration on querying efficiency in active learning. i ii Sammanfattning Många applikationer för klassificeringsmetoder kräver inte bara hög noggrannhet utan även tillförlitlig uppskattning av osäkerheten av beräknat utfall. Detta är av särskild betydelse inom områden som datorseende eller robotik, där säkerhetskritiska beslut fattas utifrån klassificeringsresultat. Medan många av de nuvarande klassi- ficeringsverktygen, i synnerhet djupa neurala nätverksarkitekturer, ger resultat när det gäller noggrannhet, tenderar de att felaktigt uppskatta strukturens osäkerhet. I detta examensarbete fokuserar vi på sannolikhetskalibrering, d.v.s. hur väl en klas- sificerares förtroende för ett resultat stämmer överens med den faktiska empiriska säkerheten. Vi studerar kalibrering ur ett teoretiskt perspektiv och kopplar det till över- och underförtroende, två begrepp som introducerades första gången i samband med aktivt lärande. Huvuddelen av arbetet är framtagandet av en ny algoritm för klassificeringskali- brering. Vi föreslår en icke-parametrisk kalibreringsmetod som, till skillnad från befintliga tillvägagångssätt, bygger på en latent Gaussisk process och som är speciellt utformad för klassificering av flera klasser. Algoritmen är inte begränsad till neu- rala nätverk utan kan tillämpas på alla klassificeringsmetoder som ger konfidens- beräkningar. Vi demonstrerar vår metods allmänt starka prestanda över olika klassifikatorer och kända datamängder från datorseende i motsats till befintliga klassificeringskalibrering- stekniker. Slutligen utvärderas effektiviteten av kalibreringen vid aktivt lärande. iii iv Acronyms CNN convolutional neural network ECE expected calibration error ELBO evidence lower bound GP Gaussian process GPS global positioning system LA Laplace approximation MC Monte Carlo MCE maximum calibration error MCMC Markov chain Monte Carlo NLL negative log-likelihood NN neural network RKHS reproducing kernel Hilbert space SVGP scalable variational Gaussian process SVM support vector machine v Notation Scalars, Vectors and Matrices θ scalar or (probability distribution) parameter x (column) vector A matrix or random variable tr A trace of the (square) matrix A Probability Theory p(x) probability density function or probability mass function p(y x) conditional density function j X D random variable X is distributed according to distribution ∼ D iid independent and identically distributed (µ, Σ) (multivariate) normal distribution with mean µ and co- N variance Σ (x µ, Σ) density of the (multivariate) normal distribution N j Cat(ρ) categorical distribution with category probabilities ρ Cat(x ρ) probability mass function of the categorical distribution j Beta(α; β) beta distribution with shape parameters α and β (µ, k) Gaussian process with mean function µ( ) and covariance GP · function k( ; θ) · · j KL[p q] Kullback-Leibler divergence of probability distributions p jj and q H(p) information-theoretic entropy of probability distribution p vi Classification and Calibration x feature vector y class label y^ class prediction z output of a classifier, either a vector of class probabilities or logits z^ confidence in prediction ECEp expected calibration error for 1 p ≤ ≤ 1 N cardinality of the training data K number of classes M number of inducing points in a scalable variational approx- imation vii viii Contents Abstract i Acronyms v Notation vi List of Tables xi List of Figures xii 1 Introduction 1 1.1 Research Question and Contribution . .2 1.2 Related Work . .3 1.3 Societal Aspects, Ethics and Sustainability . .4 1.4 Organisation . .6 2 Background 7 2.1 Classification . .7 2.2 Uncertainty Representation . .9 2.3 Measures of Uncertainty Representation . .9 2.3.1 Negative Log-Likelihood and Cross Entropy . 10 2.3.2 Calibration and Sharpness . 10 2.3.3 Over- and Underconfidence . 13 2.4 Calibration Methods . 13 2.4.1 Binary Calibration . 14 2.4.2 Multi-class Calibration . 17 2.5 Relations between Measures of Uncertainty Representation . 18 2.5.1 Calibration, Over- and Underconfidence . 18 2.5.2 Sharpness, Over- and Underconfidence . 20 3 Gaussian Process Calibration 21 3.1 Definition . 21 3.2 Inference . 22 3.2.1 Inducing Points . 23 3.2.2 Bound on the Marginal Likelihood . 24 3.2.3 Computation of the Expectation Terms . 25 3.3 Prediction . 26 ix 3.4 Online Calibration . 27 3.5 Implementation . 27 4 Experiments 29 4.1 Synthetic Data . 31 4.2 Binary Benchmark Data . 33 4.3 Multi-class Benchmark Data . 34 4.4 Active Learning . 36 5 Conclusion 39 5.1 Summary . 39 5.2 Future Work . 40 Bibliography 41 A Additional Experimental Results 47 B Multivariate Normal Distribution 51 x List of Tables 2.1 Examples of common loss functions used in classification. Loss functions allow the comparison of different classification models by scoring them using samples from (X; Y ). We list a few common loss functions for a single input - output pair (x; y)...........8 4.1 Calibration results on binary classification benchmark data sets. Average ECE1 and standard deviation of ten Monte-Carlo cross validation folds on binary benchmark data sets. Lowest calibration error per data set and classification model is indicated in bold. 35 4.2 Calibration results on multi-class classification benchmark data sets. Average ECE1 and standard deviation of ten Monte-Carlo cross validation folds on multi-class benchmark data sets. Lowest calibration error per data set and classification model is indicated in bold..................................... 37 A.1 Accuracy after calibration on binary data. Average accuracy and standard deviation of ten Monte-Carlo cross validation folds on binary benchmark data sets. 48 A.2 Accuracy after calibration on multi-class data. Average accu- racy and standard deviation of ten Monte-Carlo cross validation folds on binary benchmark data sets. 49 xi List of Figures 1.1 Example classification task in autonomous driving. Segmented scenery of Tübingen from the cityscapes data set [3] with a bounding box around an object, demonstrating an example classification task for an autonomous car. .2 1.2 Motivating example for calibration. We trained a neural network with one hidden layer on MNIST [19] and computed the classification error, the negative log-likelihood (NLL) and the expected calibration error (ECE1) over training epochs. We observe that while accuracy continues to improve on the test set, the ECE1 increases after 20 epochs. Note that this is different from classical overfitting, as the test error continues to decrease. This shows that training and calibration need to be considered independently. This can be mitigated by post- hoc calibration using our method (dashed red line). The uncertainty estimation is improved with maintained classification accuracy. .3 2.1 Illustration of the two approaches to modelling in classifica- tion. One can take one of two approaches when trying to model the latent relationship between inputs and outputs in the training data. Either one takes a discriminative approach, modelling the posterior fX;Y (y x) directly or a generative approach modelling the joint dis- j tribution fX;Y (x; y). Reprinted from [46]. .8 2.2 Illustration of calibration and sharpness. Examples of reliabil- ity diagrams and confidence histograms for a miscalibrated and not sharp classifier, a calibrated, but not sharp classifier, a classifier which is both miscalibrated and sharp and finally a calibrated and sharp classifier. The last classifier is generally the most desirable out of the four shown as its confidence estimates match its empirical accuracy and they are sufficiently close to 0 and 1 to be informative. 12
Load more

Recommended publications
  • Obtaining Well Calibrated Probabilities Using Bayesian Binning
    Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence Obtaining Well Calibrated Probabilities Using Bayesian Binning Mahdi Pakdaman Naeini1, Gregory F. Cooper1;2, and Milos Hauskrecht1;3 1Intelligent Systems Program, University of Pittsburgh, PA, USA 2Department of Biomedical Informatics, University of Pittsburgh, PA, USA 3Computer Science Department, University of Pittsburgh, PA, USA Abstract iments to perform), medicine (e.g., deciding which therapy Learning probabilistic predictive models that are well cali- to give a patient), business (e.g., making investment deci- brated is critical for many prediction and decision-making sions), and others. However, model calibration and the learn- tasks in artificial intelligence. In this paper we present a new ing of well-calibrated probabilistic models have not been non-parametric calibration method called Bayesian Binning studied in the machine learning literature as extensively as into Quantiles (BBQ) which addresses key limitations of ex- for example discriminative machine learning models that isting calibration methods. The method post processes the are built to achieve the best possible discrimination among output of a binary classification algorithm; thus, it can be classes of objects. One way to achieve a high level of model readily combined with many existing classification algo- calibration is to develop methods for learning probabilistic rithms. The method is computationally tractable, and em- models that are well-calibrated, ab initio. However, this ap- pirically accurate, as evidenced by the set of experiments proach would require one to modify the objective function reported here on both real and simulated datasets. used for learning the model and it may increase the com- putational cost of the associated optimization task.
    [Show full text]
  • Calibrated Model-Based Deep Reinforcement Learning
    Calibrated Model-Based Deep Reinforcement Learning and intuition on how to apply calibration in reinforcement Probabilistic Models This paper focuses on probabilistic learning. dynamics models T (s0|s, a) that take a current state s 2S and action a 2A, and output a probability distribution over We validate our approach on benchmarks for contextual future states s0. Web represent the output distribution over the bandits and continuous control (Li et al., 2010; Todorov next states, T (·|s, a), as a cumulative distribution function et al., 2012), as well as on a planning problem in inventory F : S![0, 1], which is defined for both discrete and management (Van Roy et al., 1997). Our results show that s,a continuous Sb. calibration consistently improves the cumulative reward and the sample complexity of model-based agents, and also en- hances their ability to balance exploration and exploitation 2.2. Calibration, Sharpness, and Proper Scoring Rules in contextual bandit settings. Most interestingly, on the A key desirable property of probabilistic forecasts is calibra- HALFCHEETAH task, our system achieves state-of-the-art tion. Intuitively, a transition model T (s0|s, a) is calibrated if performance, using 50% fewer samples than the previous whenever it assigns a probability of 0.8 to an event — such leading approach (Chua et al., 2018). Our results suggest as a state transition (s, a, s0) — thatb transition should occur that calibrated uncertainties have the potential to improve about 80% of the time. model-based reinforcement learning algorithms with mini- 0 mal computational and implementation overhead. Formally, for a discrete state space S and when s, a, s are i.i.d.
    [Show full text]
  • Offline Deep Models Calibration with Bayesian Neural Networks
    Under review as a conference paper at ICLR 2019 OFFLINE DEEP MODELS CALIBRATION WITH BAYESIAN NEURAL NETWORKS Anonymous authors Paper under double-blind review ABSTRACT In this work the authors show that Bayesian Neural Networks (BNNs) can be efficiently applied to calibrate state-of-the-art Deep Neural Networks (DNN). Our approach acts offline, i.e., it is decoupled from the training of the DNN to be cali- brated. This offline approach allow us to apply our BNN calibration to any model regardless of the limitations that the model may present during training. Note that this offline setting is also appropriate in order to deal with privacy concerns of DNN training data or implementation, among others. We show that our approach clearly outperforms other simple maximum likelihood based solutions that have recently shown very good performance, as temperature scaling (Guo et al., 2017). As an example, we reduce the Expected Calibration Error (ECE%) from 0.52 to 0.24 on CIFAR-10 and from 4.28 to 2.46 on CIFAR-100 on two Wide ResNet with 96.13% and 80.39% accuracy respectively, which are among the best results published for these tasks. Moreover, we show that our approach improves the per- formance of online methods directly applied to the DNN, e.g. Gaussian processes or Bayesian Convolutional Neural Networks. Finally, this decoupled approach allows us to apply any further improvement to the BNN without considering the computational restrictions imposed by the deep model. In this sense, this offline setting is a practical application where BNNs can be considered, which is one of the main criticisms to these techniques.
    [Show full text]
  • Discovering General-Purpose Active Learning Strategies
    1 Discovering General-Purpose Active Learning Strategies Ksenia Konyushkova, Raphael Sznitman, and Pascal Fua, Fellow, IEEE Abstract—We propose a general-purpose approach to discovering active learning (AL) strategies from data. These strategies are transferable from one domain to another and can be used in conjunction with many machine learning models. To this end, we formalize the annotation process as a Markov decision process, design universal state and action spaces and introduce a new reward function that precisely model the AL objective of minimizing the annotation cost. We seek to find an optimal (non-myopic) AL strategy using reinforcement learning. We evaluate the learned strategies on multiple unrelated domains and show that they consistently outperform state-of-the-art baselines. Index Terms—Active learning, meta-learning, Markov decision process, reinforcement learning. F 1 INTRODUCTION ODERN supervised machine learning (ML) methods is still greedy and therefore could be a subject to finding M require large annotated datasets for training purposes suboptimal solutions. and the cost of producing them can easily become pro- Our goal is therefore to devise a general-purpose data- hibitive. Active learning (AL) mitigates the problem by se- driven AL method that is non-myopic and applicable to lecting intelligently and adaptively a subset of the data to be heterogeneous datasets. To this end, we build on earlier annotated. To do so, AL typically relies on informativeness work that showed that AL problems could be naturally measures that identify unlabelled data points whose labels reworded in Markov Decision Process (MDP) terms [4], [10]. are most likely to help to improve the performance of the In a typical MDP, an agent acts in its environment.
    [Show full text]
  • On Calibration of Modern Neural Networks
    On Calibration of Modern Neural Networks Chuan Guo * 1 Geoff Pleiss * 1 Yu Sun * 1 Kilian Q. Weinberger 1 Abstract LeNet (1998) ResNet (2016) CIFAR-100 CIFAR-100 1:0 Confidence calibration – the problem of predict- ing probability estimates representative of the 0:8 true correctness likelihood – is important for 0:6 classification models in many applications. We Accuracy Accuracy discover that modern neural networks, unlike 0:4 those from a decade ago, are poorly calibrated. Avg. confidence Avg. confidence Through extensive experiments, we observe that % of Samples 0:2 depth, width, weight decay, and Batch Normal- 0:0 ization are important factors influencing calibra- 0:0 0:2 0:4 0:6 0:8 1:0 0:0 0:2 0:4 0:6 0:8 1:0 tion. We evaluate the performance of various 1:0 Outputs Outputs post-processing calibration methods on state-of- 0:8 Gap Gap the-art architectures with image and document classification datasets. Our analysis and exper- 0:6 iments not only offer insights into neural net- 0:4 work learning, but also provide a simple and Accuracy straightforward recipe for practical settings: on 0:2 Error=44.9 Error=30.6 most datasets, temperature scaling – a single- 0:0 parameter variant of Platt Scaling – is surpris- 0:0 0:2 0:4 0:6 0:8 1:0 0:0 0:2 0:4 0:6 0:8 1:0 ingly effective at calibrating predictions. Confidence Figure 1. Confidence histograms (top) and reliability diagrams (bottom) for a 5-layer LeNet (left) and a 110-layer ResNet (right) 1.
    [Show full text]
  • Calibration Techniques for Binary Classification Problems
    Calibration Techniques for Binary Classification Problems: A Comparative Analysis Alessio Martino a, Enrico De Santis b, Luca Baldini c and Antonello Rizzi d Department of Information Engineering, Electronics and Telecommunications, University of Rome ”La Sapienza”, Via Eudossiana 18, 00184 Rome, Italy Keywords: Calibration, Classification, Supervised Learning, Support Vector Machine, Probability Estimates. Abstract: Calibrating a classification system consists in transforming the output scores, which somehow state the con- fidence of the classifier regarding the predicted output, into proper probability estimates. Having a well- calibrated classifier has a non-negligible impact on many real-world applications, for example decision mak- ing systems synthesis for anomaly detection/fault prediction. In such industrial scenarios, risk assessment is certainly related to costs which must be covered. In this paper we review three state-of-the-art calibration techniques (Platt’s Scaling, Isotonic Regression and SplineCalib) and we propose three lightweight proce- dures based on a plain fitting of the reliability diagram. Computational results show that the three proposed techniques have comparable performances with respect to the three state-of-the-art approaches. 1 INTRODUCTION tion from X . The seminal example is data clustering, where aim of the learning system is to return groups Classification is one of the most important problems (clusters) of data in such a way that patterns belong- falling under the machine learning and, specifically, ing to the same cluster are more similar with respect to under the supervised learning umbrella. Generally patterns belonging to other clusters (Jain et al., 1999; speaking, it is possible to sketch three main families: Martino et al., 2017b; Martino et al., 2018b; Martino clustering, regression/function approximation, classi- et al., 2019; Di Noia et al., 2019).
    [Show full text]
  • Radar-Detection Based Classification of Moving Objects Using Machine Learning Methods
    Radar-detection based classification of moving objects using machine learning methods VICTOR NORDENMARK ADAM FORSGREN Master of Science Thesis Stockholm, Sweden 2015 Radar-detection based classification of moving objects using machine learning methods Victor Nordenmark Adam Forsgren Master of Science Thesis MMK 2015:77 MDA 520 KTH Industrial Engineering and Management Machine Design SE-100 44 STOCKHOLM ii Examensarbete MMK 2015:77 MDA 520 Radar-detection based classification of moving objects using machine learning methods Victor Nordenmark Adam Forsgren Godkänt Examinator Handledare 2015-06-17 Martin Grimheden De-Jiu Chen Uppdragsgivare Kontaktperson Scania Södertälje AB Kristian Lundh Sammanfattning I detta examensarbete undersöks möjligheten att klassificera rörliga objekt baserat på data från Dopplerradardetektioner. Slutmålet är ett system som använder billig hårdvara och utför beräkningar av låg komplexitet. Scania, företaget som har beställt detta projekt, är intresserat av användningspotentialen för ett sådant system i applikationer för autonoma fordon. Specifikt vill Scania använda klassinformationen för att lättare kunna följa rörliga objekt, vilket är en väsentlig färdighet för en autonomt körande lastbil. Objekten delas in i fyra klasser: fotgängare, cyklist, bil och lastbil. Indatan till systemet består väsentligen av en plattform med fyra stycken monopulsdopplerradars som arbetar med en vågfrekvens på 77 GHz. Ett klassificeringssystem baserat på maskininlärningskonceptet Support vector machines har skapats. Detta system har tränats och validerats på ett dataset som insamlats för projektet, innehållandes datapunkter med klassetiketter. Ett antal stödfunktioner till detta system har också skapats och testats. Klassificeraren visas kunna skilja väl på de fyra klasserna i valideringssteget. Simuleringar av det kompletta systemet gjort på inspelade loggar med radardata visar lovande resultat, även i situationer som inte finns representerade i träningsdatan.
    [Show full text]
  • Learning to Plan with Portable Symbols
    Learning to Plan with Portable Symbols Steven James 1 Benjamin Rosman 1 2 George Konidaris 3 Abstract the state space remains complex and planning continues to We present a framework for autonomously learn- be challenging. ing a portable symbolic representation that de- On the other hand, the classical planning approach is to rep- scribes a collection of low-level continuous en- resent the world using abstract symbols, with actions repre- vironments. We show that abstract representa- sented as operators that manipulate these symbols (Ghallab tions can be learned in a task-independent space et al., 2004). Such representations use only the minimal specific to the agent that, when combined with amount of state information necessary for task-level plan- problem-specific information, can be used for ning. This is appealing since it mitigates the issue of reward planning. We demonstrate knowledge transfer sparsity and admits solutions to long-horizon tasks, but in a video game domain where an agent learns raises the question of how to build the appropriate abstract portable, task-independent symbolic rules, and representation of a problem. This is often resolved manu- then learns instantiations of these rules on a per- ally, requiring substantial effort and expertise. Fortunately, task basis, reducing the number of samples re- recent work demonstrates how to learn a provably sound quired to learn a representation of a new task. symbolic representation autonomously, given only the data obtained by executing the high-level actions available to the agent (Konidaris et al., 2018). 1. Introduction A major shortcoming of that framework is the lack of On the surface, the learning and planning communities op- generalisability—an agent must relearn the appropriate sym- erate in very different paradigms.
    [Show full text]
  • Materials Genomics and Machine Learning Kevin Maik Jablonka, Daniele Ongari, Seyed Mohamad Moosavi, and Berend Smit*
    This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes. pubs.acs.org/CR Review Big-Data Science in Porous Materials: Materials Genomics and Machine Learning Kevin Maik Jablonka, Daniele Ongari, Seyed Mohamad Moosavi, and Berend Smit* Cite This: Chem. Rev. 2020, 120, 8066−8129 Read Online ACCESS Metrics & More Article Recommendations ABSTRACT: By combining metal nodes with organic linkers we can potentially synthesize millions of possible metal−organic frameworks (MOFs). The fact that we have so many materials opens many exciting avenues but also create new challenges. We simply have too many materials to be processed using conventional, brute force, methods. In this review, we show that having so many materials allows us to use big-data methods as a powerful technique to study these materials and to discover complex correlations. The first part of the review gives an introduction to the principles of big-data science. We show how to select appropriate training sets, survey approaches that are used to represent these materials in feature space, and review different learning architectures, as well as evaluation and interpretation strategies. In the second part, we review how the different approaches of machine learning have been applied to porous materials. In particular, we discuss applications in the field of gas storage and separation, the stability of these materials, their electronic properties, and their synthesis. Given the increasing interest of the scientific community in machine learning, we expect this list to rapidly expand in the coming years.
    [Show full text]
  • Measuring Calibration in Deep Learning
    Under review as a conference paper at ICLR 2020 MEASURING CALIBRATION IN DEEP LEARNING Anonymous authors Paper under double-blind review ABSTRACT Overconfidence and underconfidence in machine learning classifiers is measured by calibration: the degree to which the probabilities predicted for each class match the accuracy of the classifier on that prediction. We propose new measures for calibration, the Static Calibration Error (SCE) and Adaptive Calibration Error (ACE). These measures taks into account every prediction made by a model, in contrast to the popular Expected Calibration Error. 1 INTRODUCTION The reliability of a machine learning model’s confidence in its predictions is critical for high risk applications, such as deciding whether to trust a medical diagnosis prediction (Crowson et al., 2016; Jiang et al., 2011; Raghu et al., 2018). One mathematical formulation of the reliability of confidence is calibration (Murphy & Epstein, 1967; Dawid, 1982). Intuitively, for class predictions, calibration means that if a model assigns a class with 90% probability, that class should appear 90% of the time. Calibration is not directly measured by proper scoring rules like negative log likelihood or the quadratic loss. The reason that we need to assess the quality of calibrated probabilities is that presently, we optimize against a proper scoring rule and our models often yield uncalibrated probabilities. Recent work proposed Expected Calibration Error (ECE; Naeini et al., 2015), a measure of calibration error which has lead to a surge of works developing methods for calibrated deep neural networks (e.g., Guo et al., 2017; Kuleshov et al., 2018). In this paper, we show that ECE has numerous pathologies, and that recent calibration methods, which have been shown to successfully recalibrate models according to ECE, cannot be properly evaluated via ECE.
    [Show full text]
  • Arxiv:1808.00111V2 [Cs.LG] 14 Sep 2018
    JMLR: Workshop and Conference Proceedings 77:145{160, 2017 ACML 2017 Probability Calibration Trees Tim Leathart [email protected] Eibe Frank [email protected] Geoffrey Holmes [email protected] Department of Computer Science, University of Waikato Bernhard Pfahringer [email protected] Department of Computer Science, University of Auckland Editors: Yung-Kyun Noh and Min-Ling Zhang Abstract Obtaining accurate and well calibrated probability estimates from classifiers is useful in many applications, for example, when minimising the expected cost of classifications. Ex- isting methods of calibrating probability estimates are applied globally, ignoring the po- tential for improvements by applying a more fine-grained model. We propose probability calibration trees, a modification of logistic model trees that identifies regions of the input space in which different probability calibration models are learned to improve performance. We compare probability calibration trees to two widely used calibration methods|isotonic regression and Platt scaling|and show that our method results in lower root mean squared error on average than both methods, for estimates produced by a variety of base learners. Keywords: Probability calibration, logistic model trees, logistic regression, LogitBoost 1. Introduction In supervised classification, assuming uniform misclassification costs, it is sufficient to pre- dict the most likely class for a given test instance. However, in some applications, it is important to produce an accurate probability distribution over the classes for each exam- ple. While most classifiers can produce a probability distribution for a given test instance, these probabilities are often not well calibrated, i.e., they may not be representative of the true probability of the instance belonging to a particular class.
    [Show full text]
  • Multi-Class Probabilistic Classification Using Inductive and Cross Venn–Abers Predictors
    Proceedings of Machine Learning Research 60:1{13, 2017 Conformal and Probabilistic Prediction and Applications Multi-class probabilistic classification using inductive and cross Venn{Abers predictors Valery Manokhin [email protected] Royal Holloway, University of London, Egham, Surrey, UK Editor: Alex Gammerman, Vladimir Vovk, Zhiyuan Luo and Harris Papadopoulos Abstract Inductive (IVAP) and cross (CVAP) Venn{Abers predictors are computationally efficient algorithms for probabilistic prediction in binary classification problems. We present a new approach to multi-class probability estimation by turning IVAPs and CVAPs into multi- class probabilistic predictors. The proposed multi-class predictors are experimentally more accurate than both uncalibrated predictors and existing calibration methods. Keywords: Probabilistic classification, uncertainty quantification, calibration method, on-line compression modeling, measures of confidence 1. Introduction Multi-class classification is the problem of classifying objects into one of the more than two classes. The goal of classification is to construct a classifier which, given a new test object, will predict the class label from the set of possible k classes. In an \ordinary" classification problem, the goal is to minimize the loss function by predicting correct labels on the test set. In contrast, a probabilistic classifier outputs, for a given test object, the probability distribution over the set of k classes. Probabilistic classifiers allow to express a degree of confidence about the classification of the test object. This is useful in a range of applications and industries, including life sciences, robotics and artificial intelligence. A number of techniques are available to solve binary classification problems. In logistic regression, the dependent variable is a categorical variable indicating one of the two classes.
    [Show full text]