PIVEN: A Deep Neural Network for Prediction Intervals with Specific Value Prediction

Eli Simhayev, Gilad Katz, Lior Rokach

Keywords: Prediction Intervals, Uncertainty Estimation, Regression

Abstract: Improving the robustness of neural nets in regression tasks is key to their application in multiple domains. Deep learning-based approaches aim to achieve this goal either by improving their prediction of specific values (i.e., point prediction), or by producing prediction intervals (PIs) that quantify uncertainty. We present PIVEN, a deep neural network for producing both a PI and a prediction of specific values. Unlike previous studies, PIVEN makes no assumptions regarding data distribution inside the PI, making its point prediction more effective for various real-world problems. Benchmark experiments show that our approach produces tighter uncertainty bounds than the current state-of-the-art approach for producing PIs, while maintaining comparable performance to the state-of-the-art approach for specific value-prediction. Additional evaluation on large image datasets further support our conclusions.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

One-sentence Summary: A network architecture that provides both the value prediction and predictive interval (PI) on regression tasks.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2006.05139/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=M1rtfjhEb4