# Uncertainty Estimation and Calibration with Finite-State Probabilistic RNNs

arXiv: https://arxiv.org/abs/2011.12010 · HTML: https://mlsanity.com/paper/2011.12010/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Nov 2020.

## Built on by (most cited first, 9 of 9)

- [Calibration in Deep Learning: A Survey of the State-of-the-Art](https://mlsanity.com/paper/2308.01222/) (2023)
- [Exploring Predictive Uncertainty and Calibration in NLP: A Study on the Impact of Method & Data Scarcity](https://mlsanity.com/paper/2210.15452/) (2022)
- [Transformer Uncertainty Estimation with Hierarchical Stochastic Attention](https://mlsanity.com/paper/2112.13776/) (2021)
- [Calibrating Verbalized Probabilities for Large Language Models](https://mlsanity.com/paper/2410.06707/) (2024)
- [Calibrating a Deep Neural Network with Its Predecessors](https://mlsanity.com/paper/2302.06245/) (2023)
- [Can We Leverage Predictive Uncertainty to Detect Dataset Shift and Adversarial Examples in Android Malware Detection?](https://mlsanity.com/paper/2109.09654/) (2021)
- [Handling Long-Tail Queries with Slice-Aware Conversational Systems](https://mlsanity.com/paper/2104.13216/) (2021)
- [State-Regularized Recurrent Neural Networks to Extract Automata and Explain Predictions](https://mlsanity.com/paper/2212.05178/) (2022)
- [A Saliency-based Clustering Framework for Identifying Aberrant Predictions](https://mlsanity.com/paper/2311.06454/) (2023)

## Abstract (arXiv)

Uncertainty quantification is crucial for building reliable and trustable machine learning systems. We propose to estimate uncertainty in recurrent neural networks (RNNs) via stochastic discrete state transitions over recurrent timesteps. The uncertainty of the model can be quantified by running a prediction several times, each time sampling from the recurrent state transition distribution, leading to potentially different results if the model is uncertain. Alongside uncertainty quantification, our proposed method offers several advantages in different settings. The proposed method can (1) learn deterministic and probabilistic automata from data, (2) learn well-calibrated models on real-world classification tasks, (3) improve the performance of out-of-distribution detection, and (4) control the exploration-exploitation trade-off in reinforcement learning.