用AI解释+元学习，零样本优化模型超参数，省时省力还准确。
MetaLLMix : An XAI Aided LLM-Meta-learning Based Approach for Hyper-parameters Optimization
- 结合元学习与可解释AI，从历史实验中学习推荐最优超参数。
- 在8个医学影像数据集上5次达到最优，训练速度提升2.4-15.7倍。
- 本地部署无需昂贵API，响应时间降低99.6%-99.9%，适合医疗场景。
深度学习中的有效模型与超参数选择仍是重大挑战，常需大量专业知识和计算资源。尽管AutoML与大语言模型（LLMs）承诺自动化，但现有基于LLM的方法依赖试错与昂贵API，可解释性与泛化能力有限。本文提出MetaLLMiX，一种零样本超参数优化框架，融合元学习、可解释AI与高效LLM推理。通过利用带有SHAP解释的历史实验结果，MetaLLMiX无需额外试验即可推荐最优超参数与预训练模型。我们进一步采用LLM作为评判者，控制输出格式、准确性和完整性。在八个医学影像数据集上使用九个开源轻量级LLM的实验表明，MetaLLMiX性能媲美或优于传统HPO方法，同时大幅降低计算成本。本地部署表现优于先前基于API的方法，在8项任务中5项取得最优结果，响应时间减少99.6%-99.9%，6个数据集训练速度最快（快2.4-15.7倍），精度保持在最佳基线的1%-5%以内。
原文摘要 · Abstract (English)
Effective model and hyperparameter selection remains a major challenge in deep learning, often requiring extensive expertise and computation. While AutoML and large language models (LLMs) promise automation, current LLM-based approaches rely on trial and error and expensive APIs, which provide limited interpretability and generalizability. We propose MetaLLMiX, a zero-shot hyperparameter optimization framework combining meta-learning, explainable AI, and efficient LLM reasoning. By leveraging historical experiment outcomes with SHAP explanations, MetaLLMiX recommends optimal hyperparameters and pretrained models without additional trials. We further employ an LLM-as-judge evaluation to control output format, accuracy, and completeness. Experiments on eight medical imaging datasets using nine open-source lightweight LLMs show that MetaLLMiX achieves competitive or superior performance to traditional HPO methods while drastically reducing computational cost. Our local deployment outperforms prior API-based approaches, achieving optimal results on 5 of 8 tasks, response time reductions of 99.6-99.9%, and the fastest training times on 6 datasets (2.4-15.7x faster), maintaining accuracy within 1-5% of best-performing baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。