# i-Razor: A Differentiable Neural Input Razor for Feature Selection and Dimension Search in DNN-Based Recommender Systems

arXiv: https://arxiv.org/abs/2204.00281 · HTML: https://mlsanity.com/paper/2204.00281/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 52 references that are in the map.
- First posted on arXiv 1 Apr 2022.

## Builds on (most cited first, 17 of 17)

- [DARTS: Differentiable Architecture Search](https://mlsanity.com/paper/1806.09055/) (2018)
- [Wide & Deep Learning for Recommender Systems](https://mlsanity.com/paper/1606.07792/) (2016)
- [DeepFM: A Factorization-Machine based Neural Network for CTR Prediction](https://mlsanity.com/paper/1703.04247/) (2017)
- [Bilevel Programming for Hyperparameter Optimization and Meta-Learning](https://mlsanity.com/paper/1806.04910/) (2018)
- [xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems](https://mlsanity.com/paper/1803.05170/) (2018)
- [DARTS+: Improved Differentiable Architecture Search with Early Stopping](https://mlsanity.com/paper/1909.06035/) (2019)
- [AutoFIS: Automatic Feature Interaction Selection in Factorization Models for Click-Through Rate Prediction](https://mlsanity.com/paper/2003.11235/) (2020)
- [AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations](https://mlsanity.com/paper/2002.11252/) (2020)
- [Product-based Neural Networks for User Response Prediction over Multi-field Categorical Data](https://mlsanity.com/paper/1807.00311/) (2018)
- [Learnable Embedding Sizes for Recommender Systems](https://mlsanity.com/paper/2101.07577/) (2021)
- [AutoCross: Automatic Feature Crossing for Tabular Data in Real-World Applications](https://mlsanity.com/paper/1904.12857/) (2019)
- [AtomNAS: Fine-Grained End-to-End Neural Architecture Search](https://mlsanity.com/paper/1912.09640/) (2019)
- [Neural Input Search for Large Scale Recommendation Models](https://mlsanity.com/paper/1907.04471/) (2019)
- [Differentiable Product Quantization for End-to-End Embedding Compression](https://mlsanity.com/paper/1908.09756/) (2019)
- [AutoML for Deep Recommender Systems: A Survey](https://mlsanity.com/paper/2203.13922/) (2022)
- [Learning Multi-granular Quantized Embeddings for Large-Vocab Categorical Features in Recommender Systems](https://mlsanity.com/paper/2002.08530/) (2020)
- [Differentiable Neural Input Search for Recommender Systems](https://mlsanity.com/paper/2006.04466/) (2020)

## Built on by (most cited first, 4 of 4)

- [Experimental Analysis of Large-scale Learnable Vector Storage Compression](https://mlsanity.com/paper/2311.15578/) (2023)
- [CDR-Adapter: Learning Adapters to Dig Out More Transferring Ability for Cross-Domain Recommendation Models](https://mlsanity.com/paper/2311.02398/) (2023)
- [LeAP: Learnable Adaptive Permutation for Feature Selection in Heterogeneous and Sparse Recommender Systems](https://mlsanity.com/paper/2606.01111/) (2026)
- [VLA-ReID: Video-Level Association for Re-Identification in Multi-Object Tracking with Highly Similar Objects](https://mlsanity.com/paper/2607.17157/) (2026)

## Abstract (arXiv)

Input features play a crucial role in DNN-based recommender systems with thousands of categorical and continuous fields from users, items, contexts, and interactions. Noisy features and inappropriate embedding dimension assignments can deteriorate the performance of recommender systems and introduce unnecessary complexity in model training and online serving. Optimizing the input configuration of DNN models, including feature selection and embedding dimension assignment, has become one of the essential topics in feature engineering. However, in existing industrial practices, feature selection and dimension search are optimized sequentially, i.e., feature selection is performed first, followed by dimension search to determine the optimal dimension size for each selected feature. Such a sequential optimization mechanism increases training costs and risks generating suboptimal input configurations. To address this problem, we propose a differentiable neural input razor (i-Razor) that enables joint optimization of feature selection and dimension search. Concretely, we introduce an end-to-end differentiable model to learn the relative importance of different embedding regions of each feature. Furthermore, a flexible pruning algorithm is proposed to achieve feature filtering and dimension derivation simultaneously. Extensive experiments on two large-scale public datasets in the Click-Through-Rate (CTR) prediction task demonstrate the efficacy and superiority of i-Razor in balancing model complexity and performance.