arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.18040cs.AIcs.LG
收藏

利用深度学习自动发现积分

Automated Discovery of Integral with Deep Learning

Xiaoxin Yin

AI总结:

本文研究深度学习能否重新发现积分,展示模型可通过序列到序列预测或归纳基本积分法则,从曲线下面积概念推断函数积分。

摘要:

深度学习领域的最新进展,尤其是在大型语言模型(LLM)开发方面,已经展示了 AI 处理复杂数学问题或解决编程挑战的能力。然而,基于大量训练数据解决定义明确问题的能力,与进行科学发现这一细致过程存在显著差异。如今先进的 LLM 几乎在所有可用的人类知识上接受训练,基本上只是学习预测 token 序列。它们以类似于写文章的方式生成数学推导和编写代码,并不具备像人类科学家那样开创科学发现的能力。在本研究中,我们深入探讨了利用深度学习重新发现一个基础数学概念——积分——的潜力。通过将积分定义为曲线下面积,我们展示了 AI 如何推导给定函数的积分,例如推断出 $\int_{0}^{x} t^2 dt = \frac{x^3}{3}$ 以及 $\int_{0}^{x} ae^{bt} dt = \frac{a}{b} e^{bx} - \frac{a}{b}$。我们的实验表明,深度学习模型可以通过类似于语言翻译的 sequence-to-sequence 模型来完成推断积分的任务,也可以通过揭示积分的基本原理来完成,例如 $\int_{0}^{x} t^n dt = \frac{x^{n+1}}{n+1}$。

英文摘要:

Recent advancements in the realm of deep learning, particularly in the development of large language models (LLMs), have demonstrated AI's ability to tackle complex mathematical problems or solving programming challenges. However, the capability to solve well-defined problems based on extensive training data differs significantly from the nuanced process of making scientific discoveries. Trained on almost all human knowledge available, today's sophisticated LLMs basically learn to predict sequences of tokens. They generate mathematical derivations and write code in a similar way as writing an essay, and do not have the ability to pioneer scientific discoveries in the manner a human scientist would do. In this study we delve into the potential of using deep learning to rediscover a fundamental mathematical concept: integrals. By defining integrals as area under the curve, we illustrate how AI can deduce the integral of a given function, exemplified by inferring $\int_{0}^{x} t^2 dt = \frac{x^3}{3}$ and $\int_{0}^{x} ae^{bt} dt = \frac{a}{b} e^{bx} - \frac{a}{b}$. Our experiments show that deep learning models can approach the task of inferring integrals either through a sequence-to-sequence model, akin to language translation, or by uncovering the rudimentary principles of integration, such as $\int_{0}^{x} t^n dt = \frac{x^{n+1}}{n+1}$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑