欢迎添加小助理咨询活动详情！ 平台课程八折优惠券 超级折扣卡！课程享受七折优惠 自动驾驶的技术栈更新实在是太快了！三年前还是BEV，两年前是无图，一年期是端到端，今年是VLA和世界模型，下一步是什么呢？现在入行怎么才 能保证毕业不会被淘汰？ 其实没什么捷径，只有持续不断的更新自己的认知，这条最困难但却是最正确的路。 所以我们平台搭建了自动驾驶、具身智能和大模型三个平台，在变化中不断摸索前行的道路并反过来提升自己。别盼着稳定，要在变化里找新机会。 我们也在尽自己最大的力量推动行业的进步，如果你也想和我们一起前行，欢迎关注我们国庆节&中秋节的活动！喜逢国庆和中秋节节日，我们推出了今 年最大的优惠活动给大家，欢迎微信咨询小助理。 这一个月柱哥收到了很多的咨询，最具代表性的是：有一些深度学习的基础，怎么才能高效入门自动驾驶？ 星球优惠！新人七折续费五折 星球核心内容一览！ 自动驾驶之心 知识星球 技 最前沿的 自驾技术社区 术 f 7 P 7 5 r 6 自动驾驶VLA 世界模型 闭环仿真 扩散模型 BEV感知 --- 近40+学习路线 保持活力，持续学习 交 学术界&工业界 大佬面对面交流 4 r r VLA和WA ...

Core Viewpoint - The article discusses the revolutionary impact of the paper "Attention Is All You Need," which introduced the Transformer architecture, fundamentally changing the landscape of artificial intelligence and natural language processing [2][17]. Group 1: The Impact of the Transformer - The paper "Attention Is All You Need" has been cited 197,159 times on Google Scholar, highlighting its significant influence in the AI research community [3][26]. - The authors of the paper, known as the "Transformer Eight," have become prominent figures in the AI industry, with seven of them starting their own companies [4][24]. - The introduction of the Transformer architecture has led to a paradigm shift in AI, moving away from RNNs and enabling better handling of long-distance dependencies in language processing [17][18]. Group 2: Lukasz Kaiser's Journey - Lukasz Kaiser, one of the authors, chose to join OpenAI instead of starting a commercial venture, focusing on the pursuit of AGI [4][25]. - Kaiser has a strong academic background, holding dual master's degrees in computer science and mathematics, and has received prestigious awards for his research [7][8]. - His decision to leave a stable academic position for Google Brain in 2013 was driven by a desire for innovation in deep learning [11][12]. Group 3: The Evolution of AI Models - Kaiser and his team introduced the attention mechanism to address the limitations of RNNs, leading to the development of the Transformer model [15][17]. - The success of the Transformer has spurred a wave of entrepreneurship in the AI field, with many authors of the original paper becoming CEOs and CTOs of successful startups [24][27]. - Kaiser has been involved in the development of cutting-edge models like GPT-4 and GPT-5 at OpenAI, contributing to the forefront of AI research [27]. Group 4: Future Directions in AI - Kaiser predicts that the next phase of AI will focus on teaching models to think more deeply, emphasizing the importance of generating intermediate steps in reasoning [29]. - The upcoming ML Summit 2025 will feature Kaiser discussing the history, present, and future of reasoning models, indicating ongoing advancements in AI technology [28][30].

Core Insights - Market sentiment monitoring has become a crucial tool for corporate decision-making in the era of information explosion [1] - The selection of a professional and reliable service provider is a focal point for many companies, with key considerations including technical strength, data coverage, and service flexibility [1] Group 1: Data Monitoring Capabilities - A company's technical reserves often determine the depth of its services, exemplified by Beijing Blue Pacific Technology Co., Ltd., which has established a unique technical barrier in the big data field [3] - Blue Pacific has built a nationwide monitoring network that enables efficient collection and analysis of internet information, allowing companies to obtain market dynamics in real-time [3] - The timeliness and accuracy of data are core values of sentiment monitoring, with Blue Pacific leveraging its self-built IDC data center and numerous data detection nodes to ensure broad coverage and high precision [3] Group 2: Innovative Service Models - Blue Pacific integrates big data technology with mobile internet applications, offering customized solutions that transform complex technology into practical tools for non-technical managers [4] - The company's continuous optimization of data models enhances the analytical capabilities of vast information, helping businesses identify potential risks and uncover hidden market opportunities [4] - Blue Pacific's successful data support solutions in government evaluation demonstrate the broad applicability of its technology across various industries [4] Group 3: Sustainable Solutions - Companies should focus on whether service providers can offer sustainable solutions, with Blue Pacific maintaining sensitivity to cutting-edge technologies [4] - The company's rapid technological iteration and deep industry engagement highlight its ability to provide reliable technical support in a fast-changing market environment [4]

Core Viewpoint - Google achieved a significant victory in a 5-year antitrust case, avoiding forced breakup, with generative AI companies like OpenAI playing a crucial role in this outcome [2] Group 1: Antitrust Case and Market Impact - The U.S. government has intensified antitrust scrutiny on Silicon Valley giants, with Google being a key target, facing lawsuits since 2020 for its dominance in the search engine market [2] - A recent ruling by Judge Amit Mehta determined that Google does not need to divest its Chrome browser or Android operating system but must open more search result data to competitors and establish an antitrust technology committee [2] - Following the ruling, Google's stock surged over 8%, reflecting increased market confidence [2] Group 2: Role of Generative AI - The ruling highlighted the impact of generative AI, noting that more users are turning to AI chatbots like ChatGPT for information instead of traditional search engines, which reduces the necessity for a complete breakup of Google [2] - New AI browsers, such as Perplexity's Comet and OpenAI's upcoming browser, are redefining information retrieval through deep learning and natural language processing [3] - Despite the emergence of AI search engines, traditional search giants maintain a strong competitive advantage due to their established ecosystems and user data integration [3] Group 3: Future of Search Engines - Traditional search engines hold critical resources for the development of generative AI, including significant computing power and vast amounts of data [4] - The transition to AI-driven search is at a crossroads, with questions about whether new AI search engines can overcome cost and technical barriers, and whether traditional giants can successfully adapt to AI [4] - The ruling is considered one of the most impactful court decisions in the tech industry this century, providing a reference for other companies facing antitrust scrutiny, such as Meta, Amazon, and Apple [4]

Core Insights - AI technology is being utilized to predict the time of death for terminally ill patients, with accuracy rates improving from 40% to 80% [1] - Danish scientists have developed an AI model that can predict significant events and death dates using data from 5.96 million individuals with 280 dimensions of labels, achieving an accuracy rate of 78% [1] - Concerns have been raised regarding the potential misuse of this technology by insurance companies, leading to hesitance in making the algorithms public [1]

Quantitative Models and Construction Methods 1. Model Name: RNN-LIN - **Model Construction Idea**: Simplify the traditional GRU model by using a linear RNN structure, reducing parameter complexity while maintaining competitive performance[2][17][20] - **Model Construction Process**: - The model uses a linear RNN structure with only a forget gate and an output gate. The hidden state is updated without non-linear activation functions - Equations: $ h_{t} = f_{t} \otimes h_{t-1} + (1 - f_{t}) \otimes c_{t} $ $ y_{t} = o_{t} \otimes h_{t} $ $ f_{t} = Sigmoid(x_{t}W_{f}) $ $ o_{t} = Sigmoid(x_{t}W_{o}) $ $ c_{t} = SiLU(x_{t}W_{c}) $ - $f_{t}$: Forget gate - $o_{t}$: Output gate - $c_{t}$: Candidate state[20][21] - The model reduces parameters by approximately 50% compared to GRU[21] - **Evaluation**: The linear RNN model shows slightly weaker performance than GRU but remains competitive. Adding GLU modules improves its performance significantly[22][53] 2. Model Name: DecompGRN - **Model Construction Idea**: Extend the linear RNN by integrating cross-sectional information directly into the RNN gating mechanism, enabling simultaneous modeling of temporal and cross-sectional data[2][50] - **Model Construction Process**: - The first RNN layer outputs individual stock representations at each time step - Cross-sectional information is incorporated by grouping stocks based on market capitalization and calculating group de-meaned values - The second RNN layer combines temporal and cross-sectional information in the forget and output gates - Equations: $ h_{t} = f_{t} \otimes h_{t-1} + (1 - f_{t}) \otimes c_{t} $ $ y_{t} = o_{t} \otimes h_{t} $ $ f_{t} = Sigmoid(x_{t}W_{f}) $ $ o_{t} = Sigmoid(x_{t}W_{o}) $ $ c_{t} = SiLU(x_{t}W_{c}) $ - $f_{t}$: Forget gate - $o_{t}$: Output gate - $c_{t}$: Candidate state[50][55] - **Evaluation**: DecompGRN outperforms the GRU baseline in terms of RankIC and RankICIR while maintaining only 43% of the GRU's parameter count[74][53] --- Model Backtest Results 1. RNN-LIN - **RankIC**: - CSI All Share: 0.13 - CSI 300: 0.10 - CSI 500: 0.09 - CSI 1000: 0.12[36][37] - **RankICIR**: - CSI All Share: 1.08 - CSI 300: 0.62 - CSI 500: 0.71 - CSI 1000: 0.96[36][37] - **IC Win Rate**: - CSI All Share: 0.88 - CSI 300: 0.74 - CSI 500: 0.78 - CSI 1000: 0.86[36][37] - **Annualized Return (Top Group)**: - CSI All Share: 42.59% - CSI 300: 28.59% - CSI 500: 23.68% - CSI 1000: 32.81%[42] 2. DecompGRN - **RankIC**: - CSI All Share: 0.141 - CSI 300: 0.099 - CSI 500: 0.098 - CSI 1000: 0.127[55][58] - **RankICIR**: - CSI All Share: 1.26 - CSI 300: 0.65 - CSI 500: 0.77 - CSI 1000: 1.08[55][58] - **IC Win Rate**: - CSI All Share: 0.89 - CSI 300: 0.74 - CSI 500: 0.78 - CSI 1000: 0.88[55][58] - **Annualized Return (Top Group)**: - CSI All Share: 57.68% - CSI 300: 31.69% - CSI 500: 26.9% - CSI 1000: 40.35%[57][58] --- Index Enhancement Test Results (DecompGRN) - **Annualized Excess Return**: - CSI 300: 10.24% - CSI 500: 10.05% - CSI 1000: 19.58%[75][85] - **Tracking Error**: - CSI 300: 5.07 - CSI 500: 6.1 - CSI 1000: 6.75[75][85] - **Cumulative Excess Return (as of 2025-08-27)**: - CSI 300: 3.93% - CSI 500: 6.72% - CSI 1000: 18.26%[75][85]

你是否也有过这样的困扰？想读书却静不下心来，翻几页书视线就忍不住飘向一旁的手机；提笔想写点 什么，脑海里刚冒出零星想法，就下意识点开对话框向人工智能索要答案；有时候，甚至连一部电影都 要以倍速看完……数字时代，海量资源唾手可得，但"读不进、看不下、沉不住"成了不少人的常态。 缺乏专注的探索，难以抵达认知的深处；没有深度学习的支撑，也难以真正慎思笃行。当下，越来越多 人困扰于专注力的减弱、深度学习能力的退化。有人抱怨，生活节奏快，无时无刻不在赶时间、赶路 程、赶进度，不知不觉消磨了精气神，空闲时间只想刷刷不用过脑子的内容。也有人渴望在文化体验与 自我提升中"由浅入深"，却又陷入"知易行难"的窘境。 在碎片化的浪潮里，我们该如何守护专注力、思考力？ 找回专注力，当从改变心态做起，卸下功利的枷锁，静待花开。追求高效率本身并无不妥，一些实用性 技巧也确实可以快速掌握。但若过度追求快节奏，便容易形成对"深度"的挤压。比如，3分钟可以"看 完"一部电影，但没法欣赏镜头语言中的艺术；5分钟也可以"速读"一部名著，却难以体会隐藏于文字之 下的意蕴。只有当我们不再执着于追问"这有什么用"，不再把每一项投入都视作必须即时兑现的投 ...

Core Viewpoint - Stanford University's classic course "CS231n: Deep Learning for Computer Vision" is officially launched for Spring 2025, focusing on deep learning architectures and visual recognition tasks such as image classification, localization, and detection [1][2]. Course Overview - The course spans 10 weeks, teaching students how to implement and train neural networks while gaining insights into cutting-edge research in computer vision [3]. - At the end of the course, students will have the opportunity to train and apply neural networks with millions of parameters on real-world visual problems of their choice [4]. - Through multiple practical assignments and projects, students will acquire the necessary toolset for deep learning tasks and engineering techniques commonly used in training and fine-tuning deep neural networks [5]. Instructors - The course features four main instructors: - Fei-Fei Li: A renowned scholar and Stanford professor, known for creating the ImageNet project, which significantly advanced deep learning in computer vision [6]. - Ehsan Adeli: An assistant professor at Stanford, focusing on computer vision, computational neuroscience, and medical image analysis [6]. - Justin Johnson: An assistant professor at the University of Michigan, with research interests in computer vision and machine learning [6]. - Zane Durante: A third-year PhD student at Stanford, researching multimodal visual understanding and AI applications in healthcare [7]. Course Content - The curriculum includes topics such as: - Image classification using linear classifiers - Regularization and optimization techniques - Neural networks and backpropagation - Convolutional Neural Networks (CNNs) for image classification - Recurrent Neural Networks (RNNs) - Attention mechanisms and Transformers - Object recognition, image segmentation, and visualization - Video understanding - Large-scale distributed training - Self-supervised learning - Generative models - 3D vision - Visual and language integration - Human-centered AI [16]. Additional Resources - All 18 course videos are available for free on YouTube, with the first and last lectures delivered by Fei-Fei Li [12].

此外，项目还建立了覆盖"实时感知—精准预报—生态保护—智能防控"全周期的数据管理机制，目 前已应用于10余类海洋灾害防治业务场景。这一成果不仅在技术层面实现自主可控的海洋预报创新，更 通过多地多单位协同和数据共享机制，推动海洋数据资源的高效流通与业务化应用，为海洋防灾减灾提 供"海南智慧"和"数据样板"。 该数据集是"海南省海洋灾害综合防治能力建设项目"成果之一，该项目业主单位为海南省海洋厅， 于2025年7月竣工并通过验收，转入业务化运行阶段。 8月28日~30日，在贵阳举办的2025中国国际大数据产业博览会（数博会）上，国家数据局正式发 布高质量数据集典型案例，"海南省海洋灾害多维立体监测与智能预报预警高质量数据集"（以下简称数 据集）成功入选。 据了解，数据集聚焦台风、风暴潮、赤潮、海浪、裂流等多类海洋灾害，通过构建多维立体观测与 智能预报预警体系，有效提升海洋灾害预报的准确性、时效性与精细化水平。项目整合GPU-CPU（图 像处理器-中央处理器）异构计算、深度学习与人工智能模型，建立覆盖海南沿海的风、浪、流、风暴 潮等要素的预报模式，形成约9.6TB（太字节）高质量数据，直接服务于海洋预报警报业务。 ...

2024年12月8日，诺贝尔物理学奖得主Hinton登台，发表了题为《玻尔兹曼机》的演讲。 当时，斯德哥尔摩大学Aula Magna礼堂内座无虚席，全球目光都集聚于此。 他深入浅出地分享了，自己与John Hopfield利用神经网络，推动机器学习基础性发现的历程。 如今，Hinton这个演讲的核心内容，于8月25日正式发表在美国物理学会（APS）期刊上。 论文地址：https://journals.aps.org/rmp/pdf/10.1103/RevModPhys.97.030502 1980年代，并存两种颇具前景的梯度计算技术—— 一种是，反向传播算法，如今成为了深度学习 核心引擎，几乎无处不在。 另一种是，玻尔兹曼机器学习算法，现已不再被使用，逐渐淡出人们的视野。 这一次，Hinton的演讲重点，就是「玻尔兹曼机」。 一开场，他幽默地表示，自己打算做一件「傻」事，决定在不使用公式的情况下，向所有人解释复杂的技术概念。 霍普菲尔德网络 找到能量最低点 什么是「霍普菲尔德网络」（Hopfield Network）？ Hinton从一个简单的二进制神经元网络入手，介绍了「霍普菲尔德网络」的核心思想。 每个神 ...