# Systematic Exploration of 4-Expert Heterogeneous Mixture-of-Experts via Automated Pipeline Search

arXiv: https://arxiv.org/abs/2606.23739 · HTML: https://mlsanity.com/paper/2606.23739/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in MoE theory and gating functions, one of 12 directions in Mixture-of-experts.
- Builds on 24 of its 28 references that are in the map.
- First posted on arXiv 21 Jun 2026.

## Placed in

- Mixture-of-experts / [MoE theory and gating functions](https://mlsanity.com/moe/moe-theory-and-gating-functions/)

## Builds on (most cited first, 24 of 24)

- [DARTS: Differentiable Architecture Search](https://mlsanity.com/paper/1806.09055/) (2018)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Regularized Evolution for Image Classifier Architecture Search](https://mlsanity.com/paper/1802.01548/) (2018)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [CondConv: Conditionally Parameterized Convolutions for Efficient Inference](https://mlsanity.com/paper/1904.04971/) (2019)
- [Random Search and Reproducibility for Neural Architecture Search](https://mlsanity.com/paper/1902.07638/) (2019)
- [Hard Mixtures of Experts for Large Scale Weakly Supervised Vision](https://mlsanity.com/paper/1704.06363/) (2017)
- [Network of Experts for Large-Scale Image Categorization](https://mlsanity.com/paper/1604.06119/) (2016)
- [Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning?](https://mlsanity.com/paper/2504.06006/) (2025)
- [LEMUR Neural Network Dataset: Towards Seamless AutoML](https://mlsanity.com/paper/2504.10552/) (2025)
- [NNGPT: Rethinking AutoML with Large Language Models](https://mlsanity.com/paper/2511.20333/) (2025)
- [Preparation of Fractal-Inspired Computational Architectures for Advanced Large Language Model Analysis](https://mlsanity.com/paper/2511.07329/) (2025)
- [A Retrieval-Augmented Generation Approach to Extracting Algorithmic Logic from Neural Networks](https://mlsanity.com/paper/2512.04329/) (2025)
- [Enhancing LLM-Based Neural Network Generation: Few-Shot Prompting and Efficient Validation for Automated Architecture Design](https://mlsanity.com/paper/2512.24120/) (2025)
- [From Memorization to Creativity: LLM as a Designer of Novel Neural Architectures](https://mlsanity.com/paper/2601.02997/) (2026)
- [LLM as a Neural Architect: Controlled Generation of Image Captioning Models Under Strict API Contracts](https://mlsanity.com/paper/2512.14706/) (2025)
- [From Brute Force to Semantic Insight: Performance-Guided Data Transformation Design with LLMs](https://mlsanity.com/paper/2601.03808/) (2026)
- [Resource-Efficient Iterative LLM-Based NAS with Feedback Memory](https://mlsanity.com/paper/2603.12091/) (2026)
- [Closed-Loop LLM Discovery of Non-Standard Channel Priors in Vision Models](https://mlsanity.com/paper/2601.08517/) (2026)
- [Delta-Based Neural Architecture Search: LLM Fine-Tuning via Code Diffs](https://mlsanity.com/paper/2605.04903/) (2026)
- [MobileAgeNet: Lightweight Facial Age Estimation for Mobile Deployment](https://mlsanity.com/paper/2604.17007/) (2026)
- [Real Image Denoising with Knowledge Distillation for High-Performance Mobile NPUs](https://mlsanity.com/paper/2605.03680/) (2026)
- [Convergence Theory for Iterative LLM-Based Neural Architecture Search: A Parametric Cross-Entropy Framework with Closed-Form Proxy Reliability](https://mlsanity.com/paper/2605.30103/) (2026)

## Built on by (most cited first, 1 of 1)

- [Structure Aware Neural Architecture Search for Mixture of Experts](https://mlsanity.com/paper/2608.29817/) (2026)

## Abstract (arXiv)

We present an automated large-scale search pipeline for heterogeneous 4-Expert Mixture-of-Experts (MoE4) architectures within the LEMUR neural network dataset ecosystem. Building on a hand-crafted heterogeneous MoE reference model, we replace manual design with a deterministic code-assembly generator that systematically combines base architecture families drawn from the LEMUR database into MoE4 ensembles, each governed by a convolutional gating network with temperature scaling, mixup augmentation, and cosine-annealed learning rate scheduling. Over a 28-day campaign on an NVIDIA RTX 4090, the pipeline generated 4,463 candidate models across 197 batches, of which 1,021 were evaluated successfully. A critical finding emerged from the campaign: due to alphabetical enumeration via itertools.combinations, the entire explored search space (4.8% of the theoretical 23,751 possible 4-family combinations) is anchored to a single family, AirNet. We characterise this coverage bias precisely, identify the root cause in the generator, and propose a stratified random sampling fix. Within the AirNet anchored scope, ShuffleNet and MobileNetV3 consistently co-produce the highest-accuracy ensembles (mean accuracy up to 0.632), while FractalNet and MNASNet are identified as low-yield families warranting exclusion in future campaigns. The pipeline, analysis artefacts, and corrected generator are released as part of the open-source NNGPT project at https://github.com/ABrain-One/nn-gpt