# DLFloat: A 16-b Floating Point Format Designed for Deep Learning Training and Inference

> Research article (2019 IEEE 26th Symposium on Computer Arithmetic (ARITH), 2019) · cited 77× · AI/ML

**Wikidata**: [openalex:W2982144630](https://www.wikidata.org/wiki/openalex:W2982144630)  
**Source**: https://4ort.xyz/entity/dlfloat-a-16-b-floating-point-format-designed-for-deep-learning-training-and-inference