# Sparsely-gated Mixture-of-Expert Layers for CNN Interpretability

arXiv: https://arxiv.org/abs/2204.10598 · HTML: https://mlsanity.com/paper/2204.10598/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 22 Apr 2022.

## Placed in

- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 23 of 23)

- [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)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI](https://mlsanity.com/paper/1910.10045/) (2019)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale](https://mlsanity.com/paper/2201.05596/) (2022)
- [BASE Layers: Simplifying Training of Large, Sparse Models](https://mlsanity.com/paper/2103.16716/) (2021)
- [Learning Factored Representations in a Deep Mixture of Experts](https://mlsanity.com/paper/1312.4314/) (2013)
- [CondConv: Conditionally Parameterized Convolutions for Efficient Inference](https://mlsanity.com/paper/1904.04971/) (2019)
- [Unified Scaling Laws for Routed Language Models](https://mlsanity.com/paper/2202.01169/) (2022)
- [Hash Layers For Large Sparse Models](https://mlsanity.com/paper/2106.04426/) (2021)
- [Interpretable Convolutional Neural Networks](https://mlsanity.com/paper/1710.00935/) (2017)
- [Hard Mixtures of Experts for Large Scale Weakly Supervised Vision](https://mlsanity.com/paper/1704.06363/) (2017)
- [Go Wider Instead of Deeper](https://mlsanity.com/paper/2107.11817/) (2021)
- [Network of Experts for Large-Scale Image Categorization](https://mlsanity.com/paper/1604.06119/) (2016)
- [Residual Mixture of Experts](https://mlsanity.com/paper/2204.09636/) (2022)
- [Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety](https://mlsanity.com/paper/2104.14235/) (2021)
- [Learning Sparse Mixture of Experts for Visual Question Answering](https://mlsanity.com/paper/1909.09192/) (2019)

## Built on by (most cited first, 14 of 14)

- [Scaling Diffusion Transformers to 16 Billion Parameters](https://mlsanity.com/paper/2407.11633/) (2024)
- [Multilinear Mixture of Experts: Scalable Expert Specialization through Factorization](https://mlsanity.com/paper/2402.12550/) (2024)
- [Jointly-Learned Exit and Inference for a Dynamic Neural Network : JEI-DNN](https://mlsanity.com/paper/2310.09163/) (2023)
- [MoDE: A Mixture-of-Experts Model with Mutual Distillation among the Experts](https://mlsanity.com/paper/2402.00893/) (2024)
- [Extracting Uncertainty Estimates from Mixtures of Experts for Semantic Segmentation](https://mlsanity.com/paper/2509.04816/) (2025)
- [Towards Adversarial Robustness of Model-Level Mixture-of-Experts Architectures for Semantic Segmentation](https://mlsanity.com/paper/2412.11608/) (2024)
- [Robust Experts: the Effect of Adversarial Training on CNNs with Sparse Mixture-of-Experts Layers](https://mlsanity.com/paper/2509.05086/) (2025)
- [Toward Efficient Inference Attacks: Shadow Model Sharing via Mixture-of-Experts](https://mlsanity.com/paper/2510.13451/) (2025)
- [Graph Knowledge Distillation to Mixture of Experts](https://mlsanity.com/paper/2406.11919/) (2024)
- [The Rise of Sparse Mixture-of-Experts: A Survey from Algorithmic Foundations to Decentralized Architectures and Vertical Domain Applications](https://mlsanity.com/paper/2602.08019/) (2026)
- [Design and Behavior of Sparse Mixture-of-Experts Layers in CNN-based Semantic Segmentation](https://mlsanity.com/paper/2604.13761/) (2026)
- [Wavelet-Domain Masked Image Modeling for Color-Consistent HDR Video Reconstruction](https://mlsanity.com/paper/2602.07393/) (2026)
- [Domain-Specialized Object Detection via Model-Level Mixtures of Experts](https://mlsanity.com/paper/2604.18256/) (2026)
- [Beyond Routing: Characterising Expert Tuning and Representation in Vision Mixture-of-Experts](https://mlsanity.com/paper/2605.20610/) (2026)

## Abstract (arXiv)

Sparsely-gated Mixture of Expert (MoE) layers have been recently successfully applied for scaling large transformers, especially for language modeling tasks. An intriguing side effect of sparse MoE layers is that they convey inherent interpretability to a model via natural expert specialization. In this work, we apply sparse MoE layers to CNNs for computer vision tasks and analyze the resulting effect on model interpretability. To stabilize MoE training, we present both soft and hard constraint-based approaches. With hard constraints, the weights of certain experts are allowed to become zero, while soft constraints balance the contribution of experts with an additional auxiliary loss. As a result, soft constraints handle expert utilization better and support the expert specialization process, while hard constraints maintain more generalized experts and increase overall model performance. Our findings demonstrate that experts can implicitly focus on individual sub-domains of the input space. For example, experts trained for CIFAR-100 image classification specialize in recognizing different domains such as flowers or animals without previous data clustering. Experiments with RetinaNet and the COCO dataset further indicate that object detection experts can also specialize in detecting objects of distinct sizes.