# Towards Oracle Knowledge Distillation with Neural Architecture Search

arXiv: https://arxiv.org/abs/1911.13019 · HTML: https://mlsanity.com/paper/1911.13019/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Nov 2019.

## Built on by (most cited first, 13 of 13)

- [Learning Student-Friendly Teacher Networks for Knowledge Distillation](https://mlsanity.com/paper/2102.07650/) (2021)
- [Multi-Target Domain Adaptation with Collaborative Consistency Learning](https://mlsanity.com/paper/2106.03418/) (2021)
- [Teacher-Student Architecture for Knowledge Distillation: A Survey](https://mlsanity.com/paper/2308.04268/) (2023)
- [Teacher-Student Architecture for Knowledge Learning: A Survey](https://mlsanity.com/paper/2210.17332/) (2022)
- [AdvPicker: Effectively Leveraging Unlabeled Data via Adversarial Discriminator for Cross-Lingual NER](https://mlsanity.com/paper/2106.02300/) (2021)
- [Deeplite Neutrino: An End-to-End Framework for Constrained Deep Learning Model Optimization](https://mlsanity.com/paper/2101.04073/) (2021)
- [Real-Time Semantic Segmentation: A Brief Survey & Comparative Study in Remote Sensing](https://mlsanity.com/paper/2309.06047/) (2023)
- [Revisiting Architecture-aware Knowledge Distillation: Smaller Models and Faster Search](https://mlsanity.com/paper/2206.13130/) (2022)
- [Design Automation for Fast, Lightweight, and Effective Deep Learning Models: A Survey](https://mlsanity.com/paper/2208.10498/) (2022)
- [SalNAS: Efficient Saliency-prediction Neural Architecture Search with self-knowledge distillation](https://mlsanity.com/paper/2407.20062/) (2024)
- [Right Time to Learn:Promoting Generalization via Bio-inspired Spacing Effect in Knowledge Distillation](https://mlsanity.com/paper/2502.06192/) (2025)
- [Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study](https://mlsanity.com/paper/2305.13232/) (2023)
- [DNAD: Differentiable Neural Architecture Distillation](https://mlsanity.com/paper/2504.20080/) (2025)

## Abstract (arXiv)

We present a novel framework of knowledge distillation that is capable of learning powerful and efficient student models from ensemble teacher networks. Our approach addresses the inherent model capacity issue between teacher and student and aims to maximize benefit from teacher models during distillation by reducing their capacity gap. Specifically, we employ a neural architecture search technique to augment useful structures and operations, where the searched network is appropriate for knowledge distillation towards student models and free from sacrificing its performance by fixing the network capacity. We also introduce an oracle knowledge distillation loss to facilitate model search and distillation using an ensemble-based teacher model, where a student network is learned to imitate oracle performance of the teacher. We perform extensive experiments on the image classification datasets---CIFAR-100 and TinyImageNet---using various networks. We also show that searching for a new student model is effective in both accuracy and memory size and that the searched models often outperform their teacher models thanks to neural architecture search with oracle knowledge distillation.