# Reducing Network Agnostophobia

arXiv: https://arxiv.org/abs/1811.04110 · HTML: https://mlsanity.com/paper/1811.04110/

## Key facts

- Cited by 165 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Nov 2018.

## Built on by (most cited first, 30 of 165)

- [Generalized Out-of-Distribution Detection: A Survey](https://mlsanity.com/paper/2110.11334/) (2021)
- [ViM: Out-Of-Distribution with Virtual-logit Matching](https://mlsanity.com/paper/2203.10807/) (2022)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [Open-Set Recognition: a Good Closed-Set Classifier is All You Need?](https://mlsanity.com/paper/2110.06207/) (2021)
- [SSD: A Unified Framework for Self-Supervised Outlier Detection](https://mlsanity.com/paper/2103.12051/) (2021)
- [Towards Open Vocabulary Learning: A Survey](https://mlsanity.com/paper/2306.15880/) (2023)
- [A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges](https://mlsanity.com/paper/2110.14051/) (2021)
- [Adversarial Reciprocal Points Learning for Open Set Recognition](https://mlsanity.com/paper/2103.00953/) (2021)
- [OpenGAN: Open-Set Recognition via Open Data Generation](https://mlsanity.com/paper/2104.02939/) (2021)
- [Is Out-of-Distribution Detection Learnable?](https://mlsanity.com/paper/2210.14707/) (2022)
- [Semantically Coherent Out-of-Distribution Detection](https://mlsanity.com/paper/2108.11941/) (2021)
- [Self-Supervised Video Forensics by Audio-Visual Anomaly Detection](https://mlsanity.com/paper/2301.01767/) (2023)
- [An Improved Baseline for Sentence-level Relation Extraction](https://mlsanity.com/paper/2102.01373/) (2021)
- [Contrastive Out-of-Distribution Detection for Pretrained Transformers](https://mlsanity.com/paper/2104.08812/) (2021)
- [Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain](https://mlsanity.com/paper/2108.08487/) (2021)
- [Unknown-Aware Object Detection: Learning What You Don't Know from Videos in the Wild](https://mlsanity.com/paper/2203.03800/) (2022)
- [RMM: Reinforced Memory Management for Class-Incremental Learning](https://mlsanity.com/paper/2301.05792/) (2023)
- [Nearest Neighbor Guidance for Out-of-Distribution Detection](https://mlsanity.com/paper/2309.14888/) (2023)
- [Opening up Open-World Tracking](https://mlsanity.com/paper/2104.11221/) (2021)
- [DenseHybrid: Hybrid Anomaly Detection for Dense Open-set Recognition](https://mlsanity.com/paper/2207.02606/) (2022)
- [Outlier detection by ensembling uncertainty with negative objectness](https://mlsanity.com/paper/2402.15374/) (2024)
- [OpenAUC: Towards AUC-Oriented Open-Set Recognition](https://mlsanity.com/paper/2210.13458/) (2022)
- [Progressive Open Space Expansion for Open-Set Model Attribution](https://mlsanity.com/paper/2303.06877/) (2023)
- [Block Selection Method for Using Feature Norm in Out-of-distribution Detection](https://mlsanity.com/paper/2212.02295/) (2022)
- [Few-shot Open-set Recognition by Transformation Consistency](https://mlsanity.com/paper/2103.01537/) (2021)
- [Class-Specific Semantic Reconstruction for Open Set Recognition](https://mlsanity.com/paper/2207.02158/) (2022)
- [Learning Bounds for Open-Set Learning](https://mlsanity.com/paper/2106.15792/) (2021)
- [Understanding the Feature Norm for Out-of-Distribution Detection](https://mlsanity.com/paper/2310.05316/) (2023)
- [An Effective Baseline for Robustness to Distributional Shift](https://mlsanity.com/paper/2105.07107/) (2021)
- [A Survey on Open Set Recognition](https://mlsanity.com/paper/2109.00893/) (2021)

## Abstract (arXiv)

Agnostophobia, the fear of the unknown, can be experienced by deep learning engineers while applying their networks to real-world applications. Unfortunately, network behavior is not well defined for inputs far from a networks training set. In an uncontrolled environment, networks face many instances that are not of interest to them and have to be rejected in order to avoid a false positive. This problem has previously been tackled by researchers by either a) thresholding softmax, which by construction cannot return "none of the known classes", or b) using an additional background or garbage class. In this paper, we show that both of these approaches help, but are generally insufficient when previously unseen classes are encountered. We also introduce a new evaluation metric that focuses on comparing the performance of multiple approaches in scenarios where such unseen classes or unknowns are encountered. Our major contributions are simple yet effective Entropic Open-Set and Objectosphere losses that train networks using negative samples from some classes. These novel losses are designed to maximize entropy for unknown inputs while increasing separation in deep feature space by modifying magnitudes of known and unknown samples. Experiments on networks trained to classify classes from MNIST and CIFAR-10 show that our novel loss functions are significantly better at dealing with unknown inputs from datasets such as Devanagari, NotMNIST, CIFAR-100, and SVHN.