互鉴、互利、共赢

中新携手拓展科技合作新空间

产学协同探索数字经济新模式

第二届中新数字经济与人工智能高峰论坛

明日启幕

精彩学术报告打卡攻略请收藏→

01论坛指南

02论坛报告摘要

任福继

题目：从数字文明的视点看自然人·机器人·数字人协同共舞

Subject: From the Perspective of Digital Civilization:Human-Robot-Digital Human Collaboration

摘要：数字化的内涵从传统的信息化逐步演进到智能化、智慧化，直指数字文明。未来社会是自然人、机器人、数字人三人共舞的时代。从数字文明的视点看三人共舞需要兼具智商和情商的共享心脑，基于大模型的世界知识以及基于先进智能的情感交互是共享心脑的必备要素。数字人在未来的生产模式中发挥着重要作用。但目前的数字人离担任这个重要角色还有很长的距离。这不仅仅是数字人的外观、动作、演技，而且是数字人的对外物感受力、交互和理解力、智能并聪慧等。我们把后者称为灵气。本报告结合机器人、数字人、大模型、元宇宙、超世界的研究进展，阐述具有灵气的数字人是未来人工智能应用场景的主角。

Abstract：The essence of digitization has gradually evolved from traditional informatization to intelligence and smartification, directly pointing to digital civilization. The future society will witness the collaboration of natural humans, robots and digital humans. From the perspective of digital civilization, embracing the collaboration requires a shared heart and brain with both intelligence quotient (IQ) and emotional intelligence (EQ), of which world knowledge based on large models and affective interaction grounded in advanced intelligence are essential elements. Digital humans will play a vital role in the future production modes, there is still a long way to go for digital humans to take on the significant role as assuming the role entails not only the appearance, actions, and performance of digital humans, but also their sensitivity to external stimuli, interaction and understanding, intelligence, and wisdom. The latter was called “vitality”. This report elaborates on the fact that digital humans with “vitality” will become the protagonists of future AI application scenarios in light of the researching findings in the fields of robots, digital humans, large models, Metaverse, and “virtual world”.

安波

题目：深度强化学习用于量化交易

Subject: Deep Reinforcement Learning for Quantitative Trading

摘要：近10年来，人工智能驱动的量化交易（QT）取得重大发展。这是由于它能够即时准确地执行订单，分析并处理与金融市场相关的大量数据。传统的人工智能驱动量化交易方法基于启发式规则或金融预测来发现交易机会。然而，由于金融市场具有高波动性和噪声特性，传统方法的表现并不稳定，而且高度依赖市场状况。近日，强化学习（RL）因其在解决复杂决策问题方面的出色表现而成为量化交易（QT）任务中一种有吸引力的方法。本次讲座将讨论强化学习（RL）在量化交易（QT）中的一些最新研究进展和未来发展方向。更多信息请访问http://trademaster.ai/和https://github.com/TradeMaster-NTU/TradeMaster.

Abstract：In the last decade, we have witnessed a significant development of AI-powered quantitative trading (QT), due to its instant and accurate order execution, and capability of analyzing and processing large amount of data related to the financial market. Traditional AI-powered QT methods discover trading opportunities based on either heuristic rules or financial prediction. However, due to the high volatility and noisy nature of financial market, their performance is not stable and highly reply on the market condition. Recently, reinforcement learning (RL) becomes an appealing approach for QT tasks owing to its stellar performance on solving complex decision-making problems. This talk will discuss some recent research progress in RL for QT and future directions.  More information is at http://trademaster.ai/ and https://github.com/TradeMaster-NTU/TradeMaster.

朱达飞

题目：Web3与AI---可持续数字经济的共同基石

Subject: When Web3 meets AI for Sustainable Digital Economy

摘要：AI的繁荣带来了大量令人惊叹的创新，同时也带来了对治理问题的担忧，协同合作正日益成为未来的数字经济应用核心特征：当数据、模型和计算资源来自不同实体时，如何同时提供信任和激励从而保证数字经济的可持续性？Web3提供了一整套解决方案。在这次演讲中，我们将概述AI与Web3的潜在整合，包括挑战和核心技术组件。还将给出一些例子，以说明两者之间的结合。

Abstract：The recent boom of AI brings along with all the wonderful innovations concerns about the governance issues, in particular when the application context is bound to be increasingly collaborative in nature.  How to provide both trust and incentive when data, model and computational resources come from different entities? Web3 offers a solution. In this talk we will give an overview for the potential integration between AI and Web3, with the challenges and core technical components.  A few examples will also be given to illustrate the marriage between the two.

高阳

题目：健康医疗数据的确权与流通

Subject: Confirmation of the Ownership and Flow of Healthcare Data

摘要：健康医疗数据是数字经济中非常重要的数据要素之一。但由于其涉及到个人隐私数据，其确权与流通仍存在困难。在本报告中，探讨了通过虚拟权益方法获得个人隐私数据的处置权，通过第三方估价方法获得数据定价，以及大模型作为数据产品进行交易时的难点。

Abstract：Healthcare data are essential elements in the digital economy. However, difficulties still arise in their rights confirmation and circulation as personal privacy data are involved. The present report probes into how to obtain the disposal rights of personal privacy data through virtual equity, to get data pricing through third-party valuations, and to address the difficulties when large models are traded as data products. 

葛树志

题目：人工智能进步：前沿，经验与机遇

Subject：Advancement of Artificial Intelligence: Frontiers， Experience and Opportunities

摘要：随着深度学习的发展，以 ChatGPT/GPT4 为代表的大模型正在对人工智能领域产生深远的影响。本次讲座首先回顾人工智能发展的关键节点，介绍包括大语言模型、多模态基础模型、人工智能内容生成、选择状态空间模型等前沿技术。通过深入分析ChatGPT、LLaMA等基础模型在文本处理、语音识别、图像识别等多个模态中的应用，讲座展现这些技术如何推动AI领域的快速进步。本讲座介绍我们在可信人工智能，人工智能与机器人技术融合以及人工智能内容生成领域的最新研究成果。这些研究不仅展示了AI技术在各个领域的广泛应用潜力，也指出了未来研究的新方向。最后，讲座探讨如何有效应对由人工智能引发的科技与产业革命带来的巨大挑战，并抓住发展机会。

Abstract: With the development of deep learning, large language models represented by ChatGPT/GPT-4 are having a profound impact on the field of artificial intelligence (AI). This talk begins by reviewing the key milestones in the development of AI and introduces cutting-edge technologies, including large language models, multimodal foundational models, AI Generation Content, and Selective State Models. The tutorial provides an in-depth review of foundation models, such as ChatGPT and LLaMA, etc., on handling text processing, speech recognition and image recognition, which demonstrates how these technologies are propelling rapid advancements in AI. The tutorial presents our latest findings in the fields of trustworthy AI, the integration of AI with robotics, and AI generation contents. These studies not only showcase the wide application potential of AI technology across various domains but also highlight new directions for future research. Finally, the tutorial discusses how to effectively address the enormous challenges brought by the technological and industrial revolution triggered by AI technology and analyzes how to efficiently meet the development opportunities.

箫小奎

题目：智能时代下的数据隐私保护

Subject: Protection of Privacy and Data in the Era of Al

摘要：在数字时代，保护个人信息已变得至关重要，尤其是现今人工智能系统越来越多地处理敏感数据。本次将聚焦人工智能背景下数据隐私保护这一关键问题，以及数据隐私保护的必要措施。差分隐私已成为这一领域的黄金标准，通过确保算法输出不会泄露个体数据点来提供强有力的保护。我们将首先对差分隐私的原理和机制进行全面介绍，然后探讨数据隐私的发展格局，重点介绍前沿研究和潜在进展。

Abstract：In the digital age, safeguarding personal information has become paramount, especially as artificial intelligence systems increasingly handle sensitive data. This talk will focus on the critical issue of data privacy protection and the measures necessary to secure it in the context of AI. Differential privacy has emerged as the gold standard in this arena, offering robust protection by ensuring that the output of algorithms does not reveal individual data points. We will begin with a comprehensive introduction to the principles and mechanisms of differential privacy, and then explore the evolving landscape of data privacy, highlighting cutting-edge research and potential advancements.

陈薇

题目：智能算法安全与治理

Subject: Safety and Governance of AI Algorithms

摘要：人工智能技术在促进经济社会发展和人类文明进步的同时，也带来难以预知的风险和挑战。如何在发展人工智能的同时不断提升其安全性能、发展相关治理技术，是全人类面临的共同课题。本报告，将介绍重点实验室在算法视角下对智能算法安全的基本内涵、挑战、技术路线、治理实践等方面的认识和成果。

Abstract： While promoting economic and social development and progress in human civilization, Artificial Intelligence (AI) has also brought about unpredictable risks and challenges. How to continuously enhance AI safety performance and develop relevant governance technologies while advancing AI development has thus become a common issue for humanity. This report introduces the perception and achievements of our laboratory in terms of fundamental implications, challenges, technical routes, and governance practices of AI Safety from the perspective of algorithm.

丛高

题目：地理空间实体表征学习和城市基础模型

Subject: Geospatial Entity Representation: A Step TowardsCity Foundation Models  

摘要：报告首先介绍报告人团队的研究概述。随后，报告的其余部分将涵盖地理空间实体表征学习中的以下研究问题：1）地理空间点对象、轨迹和区域的地理空间实体表示及其应用，例如，空间关键词搜索、POI推荐、交通流量推断、区域人口估计等。2）地理空间应用的基础模型以及城市基础模型。报告的第一部分主要集中在实体表征学习以促进地理空间实体查询和分析。报告的第二部分侧重于报告人团队在地理空间实体的自监督学习方法的探索。

Abstract：The talk will begin with an overview of the recent research conducted by the speaker's team. Subsequently, the rest of the talk will cover the following research problems on geospatial entity representation: 1) Geospatial Entity Representation for point objects, trajectory, and regions and their applications, e.g., spatial keyword search, POI recommendation, speed inference, region population estimations, etc.  2) Foundation Models for Geospatial Applications and Efforts toward City Foundation Models. The first part primarily concentrates on learning representations to facilitate geospatial entity querying and analysis. The second part focuses on self-supervised learning approaches applied to geospatial entities.

郎君

题目：AI技术在游戏全球发行中的应用

Subject: Applications of AI in Global Game Publishing

摘要：在当代社会中，电子游戏在全球范围内人们的日常生活中占据了不可或缺的地位。随着全球游戏市场的快速扩张，游戏发行技术的进步在其中起到了核心的推动作用。本报告将全面展开对全球游戏市场发展趋势的阐述，并详细分析游戏发行技术的三个关键维度：用户获取、素材生成和投放优化。报告还将讨论人工智能技术在游戏发行过程中的应用及其成效。最终，报告也将针对全球游戏发行所面临的新挑战进行探讨，并提出相应的应对策略。

Abstract：In contemporary society, electronic games play an indispensable role in daily life globally. With the rapid expansion of the global gaming market, advancements in game distribution technology have played a pivotal role in this growth. This report will provide a comprehensive overview of the global gaming market trends and analyze three key aspects of game distribution technology: user acquisition, content creation, and campaign optimization. The report will also discuss the application and effectiveness of artificial intelligence technologies in the process of game distribution. Finally, it will explore the new challenges faced by global game distribution and propose corresponding strategies to address them.

孙富春

题目：数智化时代的机遇与挑战

Subject: Digital Enablement: Opportunities and Challenges

摘要：数字智能是 "数字智能 "和 "智能数字化 "的结合，利用数字技术管理人类智能，相当于实现从 "人工 "到 "智能 "的提升，使人从复杂的劳动中解放出来。本报告首先从四个阶段总结数字化时代带来的变化，进而分析数字化的机遇，包括虚拟与现实融合的社会空间，产业链，供应链与创新链的融合以及集体智慧的汇聚与应用。报告指出企业业务、运营和维护以及基础设施的数字化转型将使企业获得超越传统企业的跨时代竞争力。报告还介绍了数字智能的主要技术，包括智慧城市、工业互联网和智能生产线集成自由技术等。最后，报告探讨了数字石头面临的主要挑战和未来发展问题。

Abstract：Digital intelligence is the combination of "digital intelligence" and "intelligent digitization", which uses digital technology to manage human intelligence, equivalent to the improvement from "manual" to "intelligent", freeing people from complex labor. The report first summarizes the changes brought about by the digital age from four stages. Subsequently, the opportunities for digitalization were analyzed, including the social space of virtual and real integration, the integration of industrial and supply chain innovation chains, and the convergence and application of collective wisdom. It was pointed out that the digitalization transformation of business, operation/maintenance, and infrastructure will enable enterprises to gain intergenerational competitiveness beyond traditional enterprises. Secondly, it introduces the main technologies of digital intelligence, including smart city, industrial Internet, and intelligent production line integration free technology. Finally, the main challenges and future development of digital stone are discussed.

03电子日程册

04扫码预约直播

论坛现已在新浪、百家号、CAAI视频号、CAAI抖音号、大咖说、上直播等多个平台开启直播预约，精彩内容与您云端有约！

05论坛地址

地点：NTU North Spine LT4 ( NS4-02-34 )

抵达：打车或者坐公交车到NTU Administration Building  (50 Nanyang Ave, Singapore 639798) 步行50米即到。