Don't overfit the history -- Recursive time series data augmentation

no code implementations • 6 Jul 2022 • Amine Mohamed Aboussalah, Min-Jae Kwon, Raj G Patel, Cheng Chi, Chi-Guhn Lee

We apply RIM to diverse real world time series cases to achieve strong performance over non-augmented data on regression, classification, and reinforcement learning tasks.

Data Augmentation Time Series +1

A Deep Reinforcement Learning Framework For Column Generation

no code implementations • 3 Jun 2022 • Cheng Chi, Amine Mohamed Aboussalah, Elias B. Khalil, Juyoung Wang, Zoha Sherkat-Masoumi

Column Generation (CG) is an iterative algorithm for solving linear programs (LPs) with an extremely large number of variables (columns).

Decision Making reinforcement-learning +1

Symmetry-Augmented Representation for Time Series

no code implementations • 1 Jan 2021 • Amine Mohamed Aboussalah, Chi-Guhn Lee

We examine the hypothesis that the concept of symmetry augmentation is fundamentally linked to learning.

Management Time Series +1