Out of Distribution Detection, Generalization, and Robustness Triangle with Maximum Probability Theorem

03/23/2022
by   Amir Emad Marvasti, et al.
2

Maximum Probability Framework, powered by Maximum Probability Theorem, is a recent theoretical development, aiming to formally define probabilistic models, guiding development of objective functions, and regularization of probabilistic models. MPT uses the probability distribution that the models assume on random variables to provide an upper bound on probability of the model. We apply MPT to challenging out-of-distribution (OOD) detection problems in computer vision by incorporating MPT as a regularization scheme in training of CNNs and their energy based variants. We demonstrate the effectiveness of the proposed method on 1080 trained models, with varying hyperparameters, and conclude that MPT based regularization strategy both stabilizes and improves the generalization and robustness of base models in addition to improved OOD performance on CIFAR10, CIFAR100 and MNIST datasets.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/19/2017

Human Activity Recognition Using Robust Adaptive Privileged Probabilistic Learning

In this work, a novel method based on the learning using privileged info...
research
03/28/2018

Quantum Coupling and Strassen Theorem

We introduce a quantum generalisation of the notion of coupling in proba...
research
04/26/2015

Maximum a Posteriori Estimation by Search in Probabilistic Programs

We introduce an approximate search algorithm for fast maximum a posterio...
research
06/16/2020

Probabilistic Optimal Transport based on Collective Graphical Models

Optimal Transport (OT) is being widely used in various fields such as ma...
research
11/16/2019

Signed Input Regularization

Over-parameterized deep models usually over-fit to a given training dist...
research
12/08/2021

Information fractal dimension of mass function

Fractal plays an important role in nonlinear science. The most important...
research
05/23/2018

Learning towards Minimum Hyperspherical Energy

Neural networks are a powerful class of nonlinear functions that can be ...

Please sign up or login with your details

Forgot password? Click here to reset