# Adaptive von Mises-Fisher Likelihood Loss for Supervised Deep Time Series Hashing

arXiv: https://arxiv.org/abs/2509.19625 · HTML: https://mlsanity.com/paper/2509.19625/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 32 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Builds on (most cited first, 12 of 12)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [The UCR Time Series Archive](https://mlsanity.com/paper/1810.07758/) (2018)
- [Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019](https://mlsanity.com/paper/1911.13288/) (2019)
- [Hyperspherical Variational Auto-Encoders](https://mlsanity.com/paper/1804.00891/) (2018)
- [HashNet: Deep Learning to Hash by Continuation](https://mlsanity.com/paper/1702.00758/) (2017)
- [One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective](https://mlsanity.com/paper/2109.14449/) (2021)
- [Deep Supervised Discrete Hashing](https://mlsanity.com/paper/1705.10999/) (2017)
- [Semantic Discord: Finding Unusual Local Patterns for Time Series](https://mlsanity.com/paper/2001.11842/) (2020)
- [Efficient High-Resolution Time Series Classification via Attention Kronecker Decomposition](https://mlsanity.com/paper/2403.04882/) (2024)
- [Deep Pairwise Hashing for Cold-start Recommendation](https://mlsanity.com/paper/2011.00944/) (2020)
- [Efficient Hierarchical Contrastive Self-supervising Learning for Time Series Classification via Importance-aware Resolution Selection](https://mlsanity.com/paper/2502.10567/) (2025)

## Abstract (arXiv)

Indexing time series by creating compact binary representations is a fundamental task in time series data mining. Recently, deep learning-based hashing methods have proven effective for indexing time series based on semantic meaning rather than just raw similarity. The purpose of deep hashing is to map samples with the same semantic meaning to identical binary hash codes, enabling more efficient search and retrieval. Unlike other supervised representation learning methods, supervised deep hashing requires a discretization step to convert real-valued representations into binary codes, but this can induce significant information loss. In this paper, we propose a von Mises-Fisher (vMF) hashing loss. The proposed deep hashing model maps data to an M-dimensional hyperspherical space to effectively reduce information loss and models each data class as points following distinct vMF distributions. The designed loss aims to maximize the separation between each modeled vMF distribution to provide a better way to maximize the margin between each semantically different data sample. Experimental results show that our method outperforms existing baselines. The implementation is publicly available at https://github.com/jmpq97/vmf-hashing