A Case for Rejection in Low Resource ML Deployment

@article{White2022ACF,
  title={A Case for Rejection in Low Resource ML Deployment},
  author={J White and Pulkit Madaan and Nikhil Shenoy and Apoorv Agnihotri and Makkunda Sharma and Jigar Doshi},
  journal={ArXiv},
  year={2022},
  volume={abs/2208.06359}
}
Building reliable AI decision support systems requires a robust set of data on which to train models; both with respect to quantity and diversity. Obtaining such datasets can be difficult in resource limited settings, or for applications in early stages of deployment. Sample rejection is one way to work around this challenge, however much of the existing work in this area is ill-suited for such scenarios. This paper substantiates that position and proposes a simple solution as a proof of… 

Figures from this paper

References

SHOWING 1-10 OF 20 REFERENCES

Selective Classification for Deep Neural Networks

A method to construct a selective classifier given a trained neural network, which allows a user to set a desired risk level and the classifier rejects instances as needed, to grant the desired risk (with high probability).

Learning with Rejection

A novel framework for classification with a rejected option that consists of simultaneously learning two functions: a classifier along with a rejection function and the results of several experiments are reported showing that the kernel-based algorithms can yield a notable improvement over the best existing confidence-based rejection algorithm.

On the Foundations of Noise-free Selective Classification

This paper presents in this paper a thorough analysis of selective classification including characterizations of RC trade-offs in various interesting settings and constructs algorithms that can optimally or near optimally achieve the best possible trade-off in a controlled manner.

SelectiveNet: A Deep Neural Network with an Integrated Reject Option

This work considers the problem of selective prediction in deep neural networks, and introduces SelectiveNet, a deep neural architecture with an integrated reject option that is trained to optimize both classification (or regression) and rejection simultaneously, end-to-end.

ATRO: Adversarial Training with a Rejection Option

A classification framework with a rejection option to mitigate the performance deterioration caused by adversarial examples, and the effectiveness of ATRO is empirically confirmed using various models and real-world datasets.

Machine Learning with a Reject Option: A survey

The conditions leading to two types of rejection, ambiguity and novelty rejection are introduced and the existing architectures for models with a reject option are defined, and the standard learning strategies to train such models are described and relate traditional machine learning techniques to rejection.

Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks

The proposed ODIN method, based on the observation that using temperature scaling and adding small perturbations to the input can separate the softmax score distributions between in- and out-of-distribution images, allowing for more effective detection, consistently outperforms the baseline approach by a large margin.

Second opinion needed: communicating uncertainty in medical machine learning

An overview of different approaches to uncertainty quantification and abstention for machine learning is provided and how these techniques could improve the safety and reliability of current ML systems being used in healthcare settings are highlighted.

Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods

This paper provides an introduction to the topic of uncertainty in machine learning as well as an overview of attempts so far at handling uncertainty in general and formalizing this distinction in particular.

Machine Learning for the Developing World

Insight is provided into systematic differences between ML4D and more traditional machine learning applications and how developing world challenges can motivate the design of novel machine learning methodologies.