More Like this

1. Ordinal Hyperplane Loss

   https://doi.org/10.1109/BigData.2018.8622079  

   Vanderheyden, Bob; Xie, Ying (December 2018, 2018 IEEE International Conference on Big Data (Big Data))

   The problem of ordinal classification occurs in a large and growing number of areas. Some of the most common source and applications of ordinal data include rating scales, medical classification scales, socio-economic scales, meaningful groupings of continuous data, facial emotional intensity, facial age estimation, etc. The problem of predicting ordinal classes is typically addressed by either performing n-1 binary classification for n ordinal classes or treating ordinal classes as continuous values for regression. However, the first strategy doesn't fully utilize the ordering information of classes and the second strategy imposes a strong continuous assumption to ordinal classes. In this paper, we propose a novel loss function called Ordinal Hyperplane Loss (OHPL) that is particularly designed for data with ordinal classes. The proposal of OHPL is a significant advancement in predicting ordinal class data, since it enables deep learning techniques to be applied to the ordinal classification problem on both structured and unstructured data. By minimizing OHPL, a deep neural network learns to map data to an optimal space where the distance between points and their class centroids are minimized while a nontrivial ordinal relationship among classes are maintained. Experimental results show that deep neural network with OHPL not only outperforms the state-of-the-art alternatives on classification accuracy but also scales well to large ordinal classification problems. 

   more » « less
2. Triplet Loss-less Center Loss Sampling Strategies in Facial Expression Recognition Scenarios

   https://doi.org/10.1109/CISS56502.2023.10089734  

   Rajoli, Hossein; Lotfi, Fatemeh; Atyabi, Adham; Afghah, Fatemeh (March 2023, 57th Annual Conference on Information Sciences and Systems (CISS))
3. Leaky Hinge Loss: The First Negatively Divergent Margin-based Loss Function forClassification

   Kwon, Oh-ran; Zou, Hui (January 2023, Journal of machine learning research)
4. Simulations of inner radiation belt proton loss during geomagnetic storms: PROTON BELT LOSS

   https://doi.org/10.1002/2015JA021568  

   Engel, M. A.; Kress, B. T.; Hudson, M. K.; Selesnick, R. S. (November 2015, Journal of Geophysical Research: Space Physics)
5. Mass Loss in Evolved Stars

   https://doi.org/10.1017/S1743921323002788  

   Matthews, Lynn D. (December 2022, Proceedings of the International Astronomical Union)

   Hirota, T; Imai, H; Menten, K; Pihlstrom, Y (Ed.)

   Abstract Intense mass loss through cool, low-velocity winds is a defining characteristic of low-to-intermediate mass stars during the asymptotic giant branch (AGB) evolutionary stage. Such winds return up ∼80% of the initial stellar mass to the interstellar medium and play a major role in enriching it with dust and heavy elements. A challenge to understanding the physics underlying AGB mass loss is its dependence on an interplay between complex and highly dynamic processes, including pulsations, convective flows, shocks, magnetic fields, and opacity changes resulting from dust and molecule formation. I highlight some examples of recent advances in our understanding of late-stage stellar mass loss that are emerging from radio and (sub)millimeter observations, with a particular focus on those that resolve the surfaces and extended atmospheres of evolved stars in space, time, and frequency. 

   more » « less