Energy-Based Open-World Uncertainty Modeling for Confidence Calibration

Confidence calibration is of great importance to the reliability of decisions made by machine learning systems. However, discriminative classifiers based on deep neural networks are often criticized for producing overconfident predictions that fail to reflect the true correctness likelihood of classification accuracy. We argue that such an inability to model uncertainty is mainly caused by the closed-world nature in softmax: a model trained by the cross-entropy loss will be forced to classify input into one of K pre-defined categories with high probability. To address this problem, we for the first time propose a novel K+1-way softmax formulation, which incorporates the modeling of open-world uncertainty as the extra dimension. To unify the learning of the original K-way classification task and the extra dimension that models uncertainty, we 1) propose a novel energy-based objective function, and moreover, 2) theoretically prove that optimizing such an objective essentially forces the extra dimension to capture the marginal data distribution. Extensive experiments show that our approach, Energy-based Open-World Softmax (EOW-Softmax), is superior to existing state-of-the-art methods in improving confidence calibration.

[1]  Mohammad Norouzi,et al.  No MCMC for me: Amortized sampling for fast and stable training of energy-based models , 2021, ICLR.

[2]  Hongxia Jin,et al.  Generalized ODIN: Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Charles J. Geyer,et al.  Practical Markov Chain Monte Carlo , 1992 .

[4]  Tao Xiang,et al.  Domain Generalization in Vision: A Survey , 2021 .

[5]  Thomas G. Dietterich,et al.  Benchmarking Neural Network Robustness to Common Corruptions and Perturbations , 2018, ICLR.

[6]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[7]  Yee Whye Teh,et al.  Do Deep Generative Models Know What They Don't Know? , 2018, ICLR.

[8]  Gopinath Chennupati,et al.  On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks , 2019, NeurIPS.

[9]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[10]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Fu Jie Huang,et al.  A Tutorial on Energy-Based Learning , 2006 .

[12]  Kilian Q. Weinberger,et al.  On Calibration of Modern Neural Networks , 2017, ICML.

[13]  Andrew Y. Ng,et al.  Reading Digits in Natural Images with Unsupervised Feature Learning , 2011 .

[14]  Erik Nijkamp,et al.  Learning Non-Convergent Non-Persistent Short-Run MCMC Toward Energy-Based Model , 2019, NeurIPS.

[15]  Ying Nian Wu,et al.  Learning Energy-Based Models by Diffusion Recovery Likelihood , 2020, ICLR.

[16]  Uri Shalit,et al.  On Calibration and Out-of-domain Generalization , 2021, NeurIPS.

[17]  Terrance E. Boult,et al.  Towards Open Set Deep Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Kibok Lee,et al.  Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples , 2017, ICLR.

[19]  Mohammad Norouzi,et al.  Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One , 2019, ICLR.

[20]  Charles Blundell,et al.  Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles , 2016, NIPS.

[21]  Arthur Gretton,et al.  Generalized Energy Based Models , 2020, ICLR.

[22]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[23]  Yoshua Bengio,et al.  Better Mixing via Deep Representations , 2012, ICML.

[24]  R. Srikant,et al.  Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks , 2017, ICLR.

[25]  Yoshua Bengio,et al.  Deep Directed Generative Models with Energy-Based Probability Estimation , 2016, ArXiv.

[26]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[27]  Maya R. Gupta,et al.  To Trust Or Not To Trust A Classifier , 2018, NeurIPS.

[28]  Tomas Pfister,et al.  Distance-Based Learning from Errors for Confidence Calibration , 2020, ICLR.

[29]  Geoffrey E. Hinton,et al.  When Does Label Smoothing Help? , 2019, NeurIPS.

[30]  Jason Yosinski,et al.  Deep neural networks are easily fooled: High confidence predictions for unrecognizable images , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[31]  C. Geyer.  Markov Chain Monte Carlo Maximum Likelihood , 1991 .

[32]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[33]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[34]  Yoshua Bengio,et al.  Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling , 2020, NeurIPS.

[35]  Yee Whye Teh,et al.  Bayesian Learning via Stochastic Gradient Langevin Dynamics , 2011, ICML.

[36]  Milos Hauskrecht,et al.  Obtaining Well Calibrated Probabilities Using Bayesian Binning , 2015, AAAI.

[37]  Igor Mordatch,et al.  Implicit Generation and Modeling with Energy Based Models , 2019, NeurIPS.

[38]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[39]  Jasper Snoek,et al.  Revisiting One-vs-All Classifiers for Predictive Uncertainty and Out-of-Distribution Detection in Neural Networks , 2020, ArXiv.

[40]  Matthias Hein,et al.  Why ReLU Networks Yield High-Confidence Predictions Far Away From the Training Data and How to Mitigate the Problem , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).