AI Agent 自动发现科学新成果，速度超现有方法20倍以上。
Aster: Autonomous Scientific Discovery over 20x Faster Than Existing Methods
- 基于任务、初始代码和评估脚本，自动迭代优化程序。
- 在数学、生物、神经科学等领域达成或超越当前最优结果。
- 适合需要长周期实验的科研场景，计算成本极低。
我们提出 Aster，一个用于自主科学发现的 AI 代理，其运行速度比现有框架快 20 倍以上。给定任务、初始程序和性能评估脚本，Aster 能持续优化程序，常带来新的状态最优（SOTA）表现。其显著减少的迭代次数使原本因评估耗时过长而不可行的问题（如数小时的机器学习训练）变得可解。我们将其应用于多个领域：数学中的 Erdos 最小重叠问题、GPU 核函数优化（TriMul）、单细胞数据去噪、神经活动预测模型在 ZAPBench 上的表现，以及 NanoGPT Speedrun Competition。Aster 在所有任务中均达到或超过当前最佳水平，仅在 ZAPBench 上与人类最优解持平，但计算量不足其 1/190。Aster 可通过 asterlab.ai 的网页界面和 API 访问。
原文摘要 · Abstract (English)
We introduce Aster, an AI agent for autonomous scientific discovery capable of operating over 20 times faster than existing frameworks. Given a task, an initial program, and a script to evaluate the performance of the program, Aster iteratively improves the program, often leading to new state-of-the-art performances. Aster's significant reduction in the number of iterations required for novel discovery expands the domain of tractable problems to include tasks with long evaluation durations, such as multi-hour machine learning training runs. We applied Aster to problems in mathematics, GPU kernel engineering, biology, neuroscience, and language model training. More specifically: the Erdos minimum overlap problem, optimizing the TriMul kernel, a single-cell analysis denoising problem, training a neural activity prediction model to perform well on ZAPBench, and the NanoGPT Speedrun Competition. Aster attains SOTA results in every task, except for ZAPBench, where it matches the performance of the best human solution with less than 1/190th of the compute. Aster is accessible via a web interface and API at asterlab.ai.
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。