# The Role of Codeword-to-Class Assignments in Error-Correcting Codes: An Empirical Study

arXiv: https://arxiv.org/abs/2302.05334 · HTML: https://mlsanity.com/paper/2302.05334/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 64 references that are in the map.
- First posted on arXiv 10 Feb 2023.

## Builds on (most cited first, 15 of 15)

- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [The Implicit Bias of Gradient Descent on Separable Data](https://mlsanity.com/paper/1710.10345/) (2017)
- [Adaptive Cross-Modal Few-Shot Learning](https://mlsanity.com/paper/1902.07104/) (2019)
- [Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts](https://mlsanity.com/paper/1803.06049/) (2018)
- [Fix your classifier: the marginal value of training the last weight layer](https://mlsanity.com/paper/1801.04540/) (2018)
- [Beyond One-hot Encoding: lower dimensional target embedding](https://mlsanity.com/paper/1806.10805/) (2018)
- [DECAF: Deep Extreme Classification with Label Features](https://mlsanity.com/paper/2108.00368/) (2021)
- [LSHTC: A Benchmark for Large-Scale Text Classification](https://mlsanity.com/paper/1503.08581/) (2015)
- [Error-Correcting Output Codes with Ensemble Diversity for Robust Learning in Neural Networks](https://mlsanity.com/paper/1912.00181/) (2019)
- [Efficient Loss-Based Decoding on Graphs For Extreme Classification](https://mlsanity.com/paper/1803.03319/) (2018)
- [LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes](https://mlsanity.com/paper/2106.01487/) (2021)
- [Label Encoding for Regression Networks](https://mlsanity.com/paper/2212.01927/) (2022)
- [Log-time and Log-space Extreme Classification](https://mlsanity.com/paper/1611.01964/) (2016)
- [Integer Programming-based Error-Correcting Output Code Design for Robust Classification](https://mlsanity.com/paper/2011.00144/) (2020)
- [Error-Correcting Factorization](https://mlsanity.com/paper/1502.07976/) (2015)

## Built on by (most cited first, 2 of 2)

- [Teacher-free Latent Self-distillation and Class-separable Representations for Lightweight IoT Attack Detection](https://mlsanity.com/paper/2403.15509/) (2024)
- [Beyond one-hot encoding? Journey into compact encoding for large multi-class segmentation](https://mlsanity.com/paper/2510.00667/) (2025)

## Abstract (arXiv)

Error-correcting codes (ECC) are used to reduce multiclass classification tasks to multiple binary classification subproblems. In ECC, classes are represented by the rows of a binary matrix, corresponding to codewords in a codebook. Codebooks are commonly either predefined or problem dependent. Given predefined codebooks, codeword-to-class assignments are traditionally overlooked, and codewords are implicitly assigned to classes arbitrarily. Our paper shows that these assignments play a major role in the performance of ECC. Specifically, we examine similarity-preserving assignments, where similar codewords are assigned to similar classes. Addressing a controversy in existing literature, our extensive experiments confirm that similarity-preserving assignments induce easier subproblems and are superior to other assignment policies in terms of their generalization performance. We find that similarity-preserving assignments make predefined codebooks become problem-dependent, without altering other favorable codebook properties. Finally, we show that our findings can improve predefined codebooks dedicated to extreme classification.