Mathematical formulas are the crystallization of human wisdom in exploring the laws of nature for thousands of years. Describing the complex laws of nature with a concise mathematical formula is a constant pursuit of scientists and a great challenge for artificial intelligence. This field is called symbolic regression (SR). Symbolic regression was originally formulated as a combinatorial optimization problem, and Genetic Programming (GP) and Reinforcement Learning algorithms were used to solve it. However, GP is sensitive to hyperparameters, and these two types of algorithms are inefficient. To solve this problem, researchers treat the mapping from data to expressions as a translation problem. And the corresponding large-scale pre-trained model is introduced. However, the data and expression skeletons do not have very clear word correspondences as the two languages do. Instead, they are more like two modalities (e.g., image and text). Therefore, in this paper, we proposed MMSR. The SR problem is solved as a pure multi-modal problem, and contrastive learning is also introduced in the training process for modal alignment to facilitate later modal feature fusion. It is worth noting that to better promote the modal feature fusion, we adopt the strategy of training contrastive learning loss and other losses at the same time, which only needs one-step training, instead of training contrastive learning loss first and then training other losses. Because our experiments prove training together can make the feature extraction module and feature fusion module wearing-in better. Experimental results show that compared with multiple large-scale pre-training baselines, MMSR achieves the most advanced results on multiple mainstream datasets including SRBench. Our code is open source at https://github.com/1716757342/MMSR
SynCoBERT: Syntax-Guided Multi-Modal Contrastive Pre-Training for Code Representation
Wuhan University(武汉大学);Noah’s Ark Lab, Huawei(华为诺亚方舟实验室);Huazhong University of Science and Technology(华中科技大学);Deakin University(迪肯大学);Monash University(莫纳什大学);Yunnan University(云南大学)
Driving, Fast or Slow? Neuro-Symbolic Guidance for Motion Prediction in Multi-Modal Ground Mobility
Artificial Intelligence and Machine Learning Lab, TU Darmstadt(达姆施塔特工业大学人工智能与机器学习实验室);Honda Research Institute(本田研究所);Hessian Center for AI (hessian.AI)(黑森州人工智能中心);Centre for Cognitive Science(认知科学中心);German Center for AI (DFKI)(德国人工智能研究中心);Uncertainty in Artificial Intelligence Lab, TU Eindhoven(埃因霍温理工大学人工智能不确定性实验室)
Academy for Advanced Interdisciplinary Studies, Northeast Normal University(先进跨学科研究院,东北师范大学);Institute of Artificial Intelligence, Beihang University(人工智能研究院,北航);Institute of Applied Physics and Computational Mathematics(应用物理与计算数学研究院);National Key Laboratory of Computational Physics(计算物理国家重点实验室);School of Mathematical Sciences, East China Normal University(数学科学学院,华东师范大学);Shanghai Zhangjiang Institute of Mathematics(上海张江数学研究所)
Dept. of Mathematics and Computer Science University of Ferrara, Italy(费拉拉大学数学与计算机科学系);Dept. of Mathematical, Physical and Computer Sciences University of Parma, Italy(帕尔马大学数学、物理与计算机科学系)
Neuro-Symbolic Fusion of Wi-Fi Sensing Data for Passive Radar with Inter-Modal Knowledge Transfer
Politecnico di Milano(米兰理工大学);University of Brescia(布雷西亚大学);DEVCOM Army Research Lab(DEVCOM陆军研究实验室);University of California, Los Angeles(加州大学洛杉矶分校);Air Force Research Laboratory(空军研究实验室)
Symbolic Representation for Any-to-Any Generative Tasks
Stanford University(斯坦福大学);Fellou AI(Fellou人工智能);Fudan University(复旦大学);South China University of Technology(华南理工大学);Cornell University(康奈尔大学);University of California San Diego(加州大学圣地亚哥分校);Wuhan University(武汉大学);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Hong Kong Polytechnic University(香港理工大学);University of Southern California(南加州大学);Carnegie Mellon University(卡内基梅隆大学);LiveX AI(LiveX人工智能)
From Generality to Mastery: Composer-Style Symbolic Music Generation via Large-Scale Pre-training
Department of Mathematics University of California, San Diego(数学系加州大学圣地亚哥分校);Department of Computer Science and Engineering University of California, San Diego(计算机科学与工程系加州大学圣地亚哥分校)
miditok: A Python package for MIDI file tokenization
Sorbonne University(索邦大学);CNRS(法国国家科学研究中心);LIP6(巴黎第六大学计算机科学实验室);Aubay(Aubay公司);University of Angers(昂热大学);LERIA(LERIA实验室);ESEO(ESEO工程师学院);ERIS(ERIS实验室)
本文介绍了开源 Python 包 MidiTok,它在统一 API 下集成多种主流符号音乐分词方法,为基于语言模型的音乐生成、建模与转录等任务提供灵活、易扩展的 MIDI 文件分词工具。