O’REILLY、INTEL AI 主办
Put AI to work
2018年4月10-11日:培训
2018年4月11-13日:辅导课 & 会议
北京,中国

人工智能大会2018讲师

会有新讲师不断加入。请经常回来查看日程安排的最新变化。

过滤器

搜索讲师

Emmanuel Ameisen is an AI program director and machine learning engineer at Insight. Emmanuel has years of experience going from product ideation to effective implementations. Previously, he implemented and scaled out predictive analytics and machine learning solutions for Local Motion and Zipcar. Emmanuel holds master’s degrees in artificial intelligence, computer engineering, and management from three of France’s top schools.

Presentations

Practical considerations when shifting to using deep learning for your text data 议题 (Session)

Emmanuel Ameisen and Jeremy Karnowski share a guide for moving your company toward deep learning using a collection of NLP best practices gathered from conversations with 75+ teams from Google, Facebook, Amazon, Twitter, Salesforce, Airbnb, Capital One, Bloomberg, and others.

Yishay Carmiel is the founder of IntelligentWire, a company that develops and implements industry-leading deep learning and AI technologies for automatic speech recognition (ASR), natural language processing (NLP) and advanced voice data extraction, and the head of Spoken Labs, the strategic artificial intelligence and machine learning research arm of Spoken Communications. Yishay and his teams are currently working on bleeding-edge innovations that make the real-time customer experience a reality—at scale. Yishay has nearly 20 years’ experience as an algorithm scientist and technology leader building large-scale machine learning algorithms and serving as a deep learning expert.

Presentations

Deep learning for speech recognition and profiling 议题 (Session)

Yishay Carmiel offers an overview of neural models in speech applications, covering the dominant techniques and the elements that have contributed to the rapid progress. Yishay also looks to the future, examining which problems still remain and how far we are from solving them.

Simon Chan is a senior director of product management for Salesforce Einstein, where he oversees platform development and delivers products that empower everyone to build smarter apps with Salesforce. Simon is a product innovator and serial entrepreneur with more than 14 years of global technology management experience in London, Hong Kong, Guangzhou, Beijing, and the Bay Area. Previously, Simon was the cofounder and CEO of PredictionIO, a leading open source machine learning server (acquired by Salesforce). Simon holds a BSE in computer science from the University of Michigan, Ann Arbor, and a PhD in machine learning from University College London.

Presentations

Crossing the enterprise AI chasm 议题 (Session)

Building an end-to-end AI application in production is tremendously more complicated than simply doing algorithm modeling in a lab. Simon Chan explains how to cross the gap between AI research fantasy into real-world applications.

Roger Chen is the program cochair for the O’Reilly Artificial Intelligence Conference. Previously, he was a principal at O’Reilly AlphaTech Ventures (OATV), where he invested in and worked with early-stage startups primarily in the realm of data, machine learning, and robotics. Roger has a deep and hands-on history with technology; before he worked in venture capital, he was an engineer at Oracle, EMC, and Vicor and developed novel nanotechnology as a PhD researcher at UC Berkeley. Roger holds a BS from Boston University and a PhD from UC Berkeley, both in electrical engineering.

Presentations

周五欢迎致辞 (Friday opening welcome) 主题演讲 (Keynote)

Program chairs Ben Lorica, Jason Dai, and Roger Chen open the second day of keynotes.

周四欢迎致辞(Thursday opening welcome) 主题演讲 (Keynote)

大会日程主席 Ben Lorica、Roger Chen 与 Jason Dai致辞开始第一天主题演讲。

Jason (Jinquan) Dai is a senior principal engineer and CTO of big data technologies at Intel, where he is responsible for leading the global engineering teams (located in both Silicon Valley and Shanghai) on the development of advanced big data analytics (including distributed machine and deep learning), as well as collaborations with leading research labs (e.g., UC Berkeley AMPLab and RISELab). Jason is an internationally recognized expert on big data, cloud and distributed machine learning; he is the program co-chair of O’Reilly AI Conference in Beijing, a founding committer and PMC member of Apache Spark, and the creator of BigDL(https://github.com/intel-analytics/BigDL/), a distributed deep learning framework on Apache Spark.

Presentations

周五欢迎致辞 (Friday opening welcome) 主题演讲 (Keynote)

Program chairs Ben Lorica, Jason Dai, and Roger Chen open the second day of keynotes.

周四欢迎致辞(Thursday opening welcome) 主题演讲 (Keynote)

大会日程主席 Ben Lorica、Roger Chen 与 Jason Dai致辞开始第一天主题演讲。

王峰是Skymind的深度学习实施工程师,在软件和互联网行业从业多年,对web全栈开发和机器学习有丰富的实践经验,加入Skymind前曾经担任过全栈工程师,并且运用深度学习框架研发应用。在大数据、分布式存储、系统架构设计、机器学习等方面能够独当一面。

Presentations

用deeplearning4j框架构建神经网络分析时间序列 2天培训 (2-day Training)

在分析时间序列或者序列数据方面循环神经网络(RNN)已经被证明是非常有效的,那么在实际的案例中如何才能把循环神经网络(RNN)的优点发挥出来呐?这里将演示如何用deeplearning4j框架构建循环神经网络(RNN)来解决时间序列的问题。

Enhao Gong is a PhD student in electrical engineering at Stanford, where he is advised by John Pauly (electrical engineering) and Greg Zaharchuk (radiology), and the founder and researcher at Subtle Medical, he is pushing the performance of deep learning methods to boost the efficiency and value for medical imaging. His research focuses on applying machine learning, deep learning, and optimization for medical imaging reconstruction and processing. Recently, Enhao has been working to bridge deep learning methods with MRI reconstruction, such as enhancing image quality with deep learning and multicontrast information, solving quantitative imaging (water-fat separation, QSM, parameter mapping) using deep learning frameworks, and using generative adversarial networks (GANs) for compressed sensing MRI.

Presentations

深度学习与人工智能在神经影像中的前沿应用 议题 (Session)

人工智能与深度学习正在快速改变医疗产业发展。本讲座将介绍斯坦福的深度学习领域学者与斯坦福医院医生、教授合作研发的技术,以及如何快速地优化临床医学影像的使用。人工智能技术让医学影像的采集与处理更加快速、高效、便捷与智能。 具体技术应用包括: 1. 如何通过人工智能优化临床影像流程,优化诊断治疗规划 2. 如何通过人工智能与深度学习预测神经疾病病人的预后和疾病发展 3. 如何通过人工智能与深度学习技术加速神经影像流程 4. 如何通过人工智能与深度学习技术显著减少放射性与显影剂使用

Dr. Baining Guo is a Distinguished Scientist of Microsoft Corporation. He is Deputy Managing Director of Microsoft Research Asia, where he works on computer graphics, computer vision, and video analysis. Prior to joining Microsoft Research in 1999, he was a senior staff researcher with Intel Research in Silicon Valley. Dr. Guo got his BS degree from Beijing University and his MS and PhD degrees from Cornell University. Dr. Guo is an IEEE fellow and ACM fellow. He is also a member of Canadian Academy of Engineering.

Presentations

微软亚洲研究院的深度图像合成技术 议题 (Session)

关于微软亚洲研究院通过人工智能技术进行图像合成的最新研究概述。从把普通照片变成毕加索风格的绘画,到生成莱昂纳多·迪卡普里奥(Leonardo DiCaprio)的新图像,我们展示了深度学习所带来的新的可能性。

Yufeng Guo is a developer advocate for the Google Cloud Platform, where he is trying to make machine learning more understandable and usable for all. He enjoys hearing about new and interesting applications of machine learning, so be sure to share your use case with him on Twitter.

Presentations

Getting up and running with TensorFlow 教学辅导课 (Tutorial)

Yufeng Guo walks you through training a machine learning system using popular open source library TensorFlow, starting from conceptual overviews and building all the way up to complex classifiers. Along the way, you'll gain insight into deep learning and how it can apply to complex problems in science and industry.

Kristian Hammond is Narrative Science’s chief scientist and a professor of computer science and journalism at Northwestern University. Previously, Kris founded the University of Chicago’s Artificial Intelligence Laboratory. His research has been primarily focused on artificial intelligence, machine-generated content, and context-driven information systems. He currently sits on a United Nations policy committee run by the United Nations Institute for Disarmament Research (UNIDIR). Kris was also named 2014 innovator of the year by the Best in Biz Awards. He holds a PhD from Yale.

Presentations

Bringing AI into the enterprise 教学辅导课 (Tutorial)

Even as AI technologies move into common use, many enterprise decision makers remain baffled about what the different technologies actually do and how they can be integrated into their businesses. Rather than focusing on the technologies alone, Kristian Hammond provides a practical framework for understanding your role in problem solving and decision making.

Some are cognitive scientists; others are computer scientists and engineers. Mark Hammond is a cognitive entrepreneur bringing together both fields along with business acumen. He has a deep passion for understanding how the mind works, combined with an understanding of own human nature, and turns that knowledge into beneficial applied technology. As the founder and CEO of Bonsai, Mark is enabling AI for everyone. Mark has been programming since the first grade and started working at Microsoft as an intern and contractor while still in high school. He has held positions at Microsoft and numerous startups and in academia, including turns at Numenta and in the Yale Neuroscience Department. He holds a degree in computation and neural systems from Caltech.

Presentations

Deep reinforcement learning’s killer app: Intelligent control in real-world systems 议题 (Session)

Mark Hammond dives into two case studies highlighting how deep reinforcement learning can be applied to real-world industrial applications.

Get your hard hat: Intelligent industrial systems with deep reinforcement learning 主题演讲 (Keynote)

Mark Hammond explores a wide breadth of real-world applications of deep reinforcement learning, including robotics, manufacturing, energy, and supply chain. Mark also shares best practices and tips for building and deploying these systems, highlighting the unique requirements and challenges of industrial AI applications.

郝井华,现任美团点评研究员,配送算法策略架构师。博士毕业于清华大学自动化系,毕业后留校工作,研究人工智能技术在制造、物流、公共服务领域中的应用。

郝井华是国内运筹优化、智能调度领域的前沿专家,在业务分析、数学建模、系统仿真和优化上有丰富的研究经验,并在如何深度结合运筹优化和机器学习技术以解决实际业务难题上进行了大量实践,发表学术论文30多篇,发明专利20项,曾获国家科技进步奖和中国物流科技进步奖等荣誉。

目前致力于解决即时配送业务中的订单分配、路径规划、动态定价等难题,期望借助人工智能技术高效实现『最后一公里』配送业务的动态最优化,更好地服务亿万用户的配送需求。

Presentations

即时配送调度中的人工智能 议题 (Session)

近两年外卖行业发展迅速,美团外卖每日超过1600万订单,线下有50万名骑手每天奔波在大街小巷进行配送,是全球最大的外卖平台。如何使数据巨大的骑手配送得更有效率,减少空驶?如何让用户更早地享受到美食,减少超时率?这是一个强随机环境下的大规模复杂优化问题。本次分享将介绍美团配送在运用大数据、机器学习和运筹优化技术解决即时配送业务难题、利用 AI 技术来取代人工上的若干进展和探索,帮助大家了解这一技术领域的进展和挑战。

Catherine Havasi is cofounder and CEO of Luminoso, an AI-based natural language understanding company located in Cambridge, MA. Luminoso’s solutions are based upon nearly a decade of Catherine’s research at the MIT Media Lab on applying natural language processing and machine learning to improve text analytics. As a woman in machine learning (especially embeddings/deep learning), as well as a women tech CEO, she offers a unique (and rare) perspective on the field. Catherine directs the Open Mind Common Sense Project, one of the largest common sense knowledge bases in the world, which she cofounded alongside Marvin Minsky and Push Singh in 1999.

Presentations

Transfer learning and the future of AI 议题 (Session)

The next frontier in AI is transfer learning, which enables computers to apply what they’ve learned in one scenario to new situations, making AI-based systems far more powerful, reusable, and flexible. But is it ready for enterprise deployment, and if so, how can it be applied to solve business problems? Join Catherine Havasi to find out.

Hsiao-Wuen Hon is corporate vice president of Microsoft, chairman of Microsoft’s Asia-Pacific R&D Group, and managing director of Microsoft Research Asia, where he drives Microsoft’s strategy for research and development activities in the Asia-Pacific region, as well as collaborations with academia. Hsiao-Wuen has been with Microsoft since 1995. Previously, he founded and managed Microsoft’s Search Technology Center and led the development of Microsoft’s search products (Bing) in Asia-Pacific. Prior to joining Microsoft Research Asia, he was the founding member and architect of the Natural Interactive Services Division at Microsoft Corporation. An IEEE fellow and a distinguished scientist of Microsoft, Hsiao-Wuen is an internationally recognized expert in speech technology. He has published more than 100 technical papers in international journals and at conferences.

Presentations

智能简史 主题演讲 (Keynote)

人工智能已经引发了众多关注和讨论,而关于人类智能和人工智能孰优孰劣的辩论也不断升温。在这个主题演讲中,洪小文博士将介绍人工智能(AI)以及人类智能(HI)的历史。从历史的维度,以深刻的洞察,阐述AI和HI是如何彼此交织并共同进化的,并预示AI和HI可能的未来。

Yonggang Hu is Distinguished Engineer, Chief Architect at Platform Computing, IBM. He has been working on distributed computing, grid, cloud and big data for the past 20 years. Before joining Platform Computing, Yonggang was Vice President and Application Architect at JPMorgan Chase focusing on computational analytics and application infrastructure. Yonggang holds MS in Computer Science from Peking University and MBA from Cornell University.

Presentations

基于Apache Spark的弹性调度在GPU/CPU异构环境中的深度学习应用 议题 (Session)

深度学习技术是从海量数据集中构建人工智能的关键技术。将Apache Spark与诸如Caffe, MxNet等深度学习框架的集成之后,可以使得后者的学习阶段能够大规模并行化,但在企业部署中会面临很多问题。我们将会分享我们在使用Apache Spark进行深度学习,特别是使用GPU的深度学习的方法以及相应的认知计算实际案例。

Xianyan Jia is a software engineer at Intel, where she’s responsible for developing deep learning and machine learning algorithms and pipelines. She is also a contributor to BigDL, a distributed deep learning framework on Apache Spark.

Presentations

基于BigDL的超大规模图像处理在京东的实践 议题 (Session)

BigDL(基于Apache Spark的大数据分布式的深度学习框架)为大规模图像处理提供了丰富的端到端支持。我们将介绍如何使用BigDL搭建灵活性和高可扩展性的端到端深度学习应用程序。我们还将分享我们在京东构建大规模图像特征提取流水线的经验。

Arthur Juliani is a machine learning engineer at Unity Technologies. A researcher working at the intersection of cognitive neuroscience and deep learning, Arthur is currently working toward a PhD at the University of Oregon.

Presentations

Deep reinforcement learning tutorial 教学辅导课 (Tutorial)

Recently, computers have been able to learn to play Atari games, Go, and first-person shooters at a superhuman level. Underlying all these accomplishments is deep reinforcement learning. Arthur Juliani offers a deep dive into reinforcement learning, from the basics using lookup tables and GridWorld all the way to solving complex 3D tasks with deep neural networks.

Sangkeun Jung is a researcher at SK Telecom, where he focuses on natural language interfaces. He is the main developer for the NUGU AI speaker. Sangkeun has 13 years of experience in AI research and engineering. He holds a PhD in computer science and engineering.

Presentations

Building a commercial natural language understanding system 议题 (Session)

Natural language understanding is a core technology for building natural interfaces such as AI speakers, chatbots, and smartphones. Sangkeun Jung offers an overview of a spoken dialog system and recently launched AI speaker, NUGU, and shares lessons learned building a commercially efficient and sustainable natural language understanding system.

Jeremy Karnowski is a data scientist at Insight Data Science, an education startup that helps top academic data scientists transition into industry data science roles. His background is in using data science, machine learning, signal processing, deep learning, and computer vision for audio and video analysis.

Presentations

Practical considerations when shifting to using deep learning for your text data 议题 (Session)

Emmanuel Ameisen and Jeremy Karnowski share a guide for moving your company toward deep learning using a collection of NLP best practices gathered from conversations with 75+ teams from Google, Facebook, Amazon, Twitter, Salesforce, Airbnb, Capital One, Bloomberg, and others.

Feng Kuan is an architect at IBM Canada focusing on Spark and AI development.

Presentations

基于Apache Spark的弹性调度在GPU/CPU异构环境中的深度学习应用 议题 (Session)

深度学习技术是从海量数据集中构建人工智能的关键技术。将Apache Spark与诸如Caffe, MxNet等深度学习框架的集成之后,可以使得后者的学习阶段能够大规模并行化,但在企业部署中会面临很多问题。我们将会分享我们在使用Apache Spark进行深度学习,特别是使用GPU的深度学习的方法以及相应的认知计算实际案例。

Danny Lange is vice president of AI and machine learning at Unity Technologies, where he leads multiple initiatives around applied artificial intelligence. Previously, Danny was head of machine learning at Uber, where he led the efforts to build a highly scalable machine learning platform to support all parts of Uber’s business, from the Uber app to self-driving cars; general manager of Amazon Machine Learning, where he provided internal teams with access to machine intelligence and launched an AWS product that offers machine learning as a cloud service to the public; principal development manager at Microsoft, where he led a product team focused on large-scale machine learning for big data; CTO of General Magic, Inc.; and founder of his own company, Vocomo Software, where he worked on General Motor’s OnStar Virtual Advisor, one of the largest deployments of an intelligent personal assistant until Siri. Danny started his career as a computer scientist at IBM Research. He is a member of ACM and IEEE Computer Society and has numerous patents to his credit. Danny holds an MS and PhD in computer science from the Technical University of Denmark.

Presentations

Conducting machine learning research within custom-made 3D game environments 议题 (Session)

Danny Lange demonstrates the role games can play in driving the development of reinforcement learning algorithms. Danny uses the Unity Engine with the ML-Agents toolkit as an example of how dynamic 3D game environments can be utilized for machine learning research.

Democratizing deep reinforcement learning 主题演讲 (Keynote)

Danny Lange offers an overview of deep reinforcement learning, an exciting new chapter in AI’s history that is changing the way we develop and test learning algorithms that can later be used in real life.

Li Erran Li is a senior research scientist in Uber’s Advanced Technologies Group and an adjunct professor in the Computer Science Department at Columbia University. His current research interests include AI, computer vision, and machine learning algorithms and systems. He is an IEEE fellow and an ACM Fellow. He holds a PhD in computer science from Cornell University, where he was advised by Joseph Halpern.

Presentations

用于自动驾驶的机器学习 教学辅导课 (Tutorial)

尽管最近人工智能等领域取得了很多的进展,但自动驾驶里的主要问题(不管是基础研究还是工程应用上的挑战)离完全被解决还有很大的距离。Erran Li将会探索自动驾驶所用的机器学习的基础,并讨论目前相关工作的进展。

用于自动驾驶的机器学习:近期的进步和未来的挑战 议题 (Session)

深度增强学习已经让人工智能体在很多挑战性的领域可以取得超越人类的表现,例如玩Atari的游戏以及下围棋。这一方法还具有能显著地推进自动驾驶的潜力。Erran Li将会讨论近期在模仿学习方面(例如infoGAIL)、策略梯度法和层次增强学习(例如option-critic架构)等方面的进步,以及它们在自动驾驶方面的应用。Erran接着还会介绍在这个领域需要关注的剩余的挑战。

Li joined ESRI as a product engineer in 2015. Prior to that, she worked in IBM China Research Lab. She obtained her PHD from the State University of New York at Buffalo.

Presentations

用深度学习给地图换新颜 议题 (Session)

制图学是一个历史悠久的学科。古希腊地理学家C.托勒密的《地理学指南》就是一部地图制图学著作。托勒密认为地理学就是“以线画形式描绘地球上所有迄今已知的部分及其附属的东西”。几百年以来,地图学领域都没有重大突破。 深度学习作为一个新的技术已经渗透到了各个行业。带来了各种各种的技术革新。本讲座就是探讨如何用深度学习来给地图换装。然后展示一些用深度学习技术给地图换装的结果。并讨论,深度学习在制图领域的应用。

李力耘,百度美国研发中心高级架构师。本科毕业于清华大学电子工程系,后获得美国纽约大学计算机专业博士学位。 百度美国无人车研发团队的创始核心成员,目前在百度无人车部门负责无人车行为预测系统和智能决策规划系统的整体架构及算法优化。拥有多项国际专利,其中已递交三十余项无人车决策预测相关专利申请,并且著有两本无人车方向的技术书籍"第一本无人驾驶技术书"(电子工业出版社出版)以及"Creating Autonomous Vehicle Systems"(MorganClaypool出版).

Presentations

自动驾驶系统中的人工智能: Artificial intelligence in autonomous vehicle systems 议题 (Session)

尽管人工智能技术已经在诸如计算机视觉和自然语言处理等领域获得了巨大的成功,如何在自动驾驶系统中有效地利用AI的能力仍然是一个很大的挑战。我们将以"Apollo"这一百度的开源无人驾驶平台系统做为基准和样例, 深入讨论并且分享在搭建智能的无人驾驶系统各个方面利用AI技术的实践和经验。通过讲解Apollo无人驾驶系统背后的设计理念以及各个功能模块,我们将分享并展示AI技术在Apollo无人驾驶系统中各方面的应用, 包括环境感知,行为预测,行为决策,以及控制规划等。同时我们将结合Apollo系统中的端到端学习实践,探讨AI技术在未来无人驾驶系统中更好的应用场景。

Tianhui Michael Li is the founder and CEO of the Data Incubator. Michael has worked as a data scientist lead at Foursquare, a quant at D.E. Shaw and JPMorgan, and a rocket scientist at NASA. At Foursquare, Michael discovered that his favorite part of the job was teaching and mentoring smart people about data science. He decided to build a startup that lets him focus on what he really loves. He did his PhD at Princeton as a Hertz fellow and read Part III Maths at Cambridge as a Marshall scholar.

Presentations

用TensorFlow进行深度学习 2天培训 (2-day Training)

TensorFlow是一个流行的深度学习的工具。我们会介绍TensorFlow的流程图、学习使用它的Python API,并展示它的用处。我们会从简单的机器学习算法开始,然后实现神经网络。我们还会讨论一些真实的深度学习的应用,包括机器视觉、文本处理和生成型网络。

Zhichao Li is a senior software engineer at Intel focused on distributed machine learning, especially large-scale analytical applications and infrastructure on Spark. He’s also an active contributor to Spark. Previously, Zhichao worked in Morgan Stanley’s FX Department.

Presentations

基于Apache Spark及BigDL构建高级数据分析 教学辅导课 (Tutorial)

从这个教学课程里,学员将会学到如何应用深度学习(最先进的机器学习技术)到他们的Apache Spark驱动的大数据工作任务里

林嘉薇是Skymind的公司讲师、深度学习工程师。毕业于汉阳大学计算机科学专业硕士毕业,在校期间的主研方向为机器学习,机器视觉和图像处理。期间作为该校的教学助理,曾为该校提供图像处理和数值分析教学课程。嘉薇还曾担任过系统工程师,为英特尔提供计算机视觉技术方案。

Presentations

用deeplearning4j框架构建神经网络分析时间序列 2天培训 (2-day Training)

在分析时间序列或者序列数据方面循环神经网络(RNN)已经被证明是非常有效的,那么在实际的案例中如何才能把循环神经网络(RNN)的优点发挥出来呐?这里将演示如何用deeplearning4j框架构建循环神经网络(RNN)来解决时间序列的问题。

林晖
英语流利说联合创始人、首席科学家
林晖博士曾任 Google 美国总部研究科学家,主要研究领域包括语音识别、自然语言处理、机器学习以及大数据挖掘。林博士在顶级杂志发表论文超过 30 篇。林晖博士毕业于美国华盛顿大学电子工程系,早年曾在清华大学获得电子工程专业硕士、学士学位。

Presentations

深度学习在智能教育中的应用 议题 (Session)

教育的个性化和高效率离不开智能化。本次演讲将结合“流利说”在过去5年的实践,从问题定义、数据获取、算法设计、模型优化等方面介绍如何将深度学习运用于语音识别、知识跟踪、以及自然语言处理等领域。实验结果显示,搭载这些智能技术的学习产品能将学习效率提升三倍。

刘俊峰,是IBM Platform Computing的软件架构师,关注于大数据平台的设计和实现,成功的向多个重要客户提供技术解决方案。

Presentations

基于Apache Spark的弹性调度在GPU/CPU异构环境中的深度学习应用 议题 (Session)

深度学习技术是从海量数据集中构建人工智能的关键技术。将Apache Spark与诸如Caffe, MxNet等深度学习框架的集成之后,可以使得后者的学习阶段能够大规模并行化,但在企业部署中会面临很多问题。我们将会分享我们在使用Apache Spark进行深度学习,特别是使用GPU的深度学习的方法以及相应的认知计算实际案例。

Shaoshan Liu is the cofounder and chairman of PerceptIn, a company working on developing a next-generation robotics platform. Previously, he worked on autonomous driving and deep learning infrastructure at Baidu USA. Shaoshan holds a PhD in computer engineering from the University of California, Irvine.

刘少山,PerceptIn联合创始人,董事长。加州大学欧文分校计算机博士,研究方向包括人工智能,无人驾驶,机器人,系统软件与异构计算。 PerceptIn专注于开发智能机器人系统,包括家用机器人,工业机器人,以及无人驾驶。 在创立PerceptIn之前,刘少山博士在人工智能以及系统方向有超过十年的研发经验,其经历包括英特尔研究院(INTEL RESEARCH),法国国家信息与自动化研究所(INRIA),微软研究院(MICROSOFT RESEARCH),微(MICROSOFT), 领英(LinkedIn),以及百度美国研究院 (Baidu USA)。

Presentations

高性价比AI产品在IoT设备上的实现 议题 (Session)

通过深度学习技术,物联网(IoT)设备能够得以解析非结构化的多媒体数据,智能地响应用户和环境事件,但是却伴随着苛刻的性能和功耗要求。我们探讨了两种方式以便将深度学习和低功耗的物联网设备成功整合。

刘铁岩博士,微软亚洲研究院副院长、首席研究员,领导机器学习和人工智能方向的研究工作。同时他也是美国卡内基-梅隆大学(CMU)客座教授、英国诺丁汉大学荣誉教授、中国科技大学、中山大学、南开大学的博士生导师。刘博士的先锋性工作促进了机器学习与信息检索之间的融合,被国际学术界公认为“排序学习”领域的代表人物,他在该领域的学术论文已被引用万余次,并受Springer出版社之邀撰写了该领域的首部学术专著(并成为Springer计算机领域华人作者十大畅销书之一)。近年来,刘博士在博弈机器学习、深度学习、分布式机器学习等方面也颇有建树,他的研究工作多次获得最佳论文奖、最高引用论文奖、研究突破奖、最佳研究团队奖;被广泛应用在微软的产品和在线服务中、并通过微软认知工具包(CNTK)、微软分布式机器学习工具包(DMTK)、微软图引擎(Graph Engine)等项目开源。他曾受邀担任了包括SIGIR、WWW、KDD、ICML、NIPS、AAAI、ACL在内的顶级国际会议的组委会主席、程序委员会主席、或领域主席;以及包括ACM TOIS、ACM TWEB、Neurocomputing在内的国际期刊副主编。他是国际电子电气工程师学会(IEEE)院士,美国计算机学会(ACM)杰出会员,中文信息学会(CIPS)信息检索专委会副主任,中国云体系产业创新战略联盟常务理事。

Presentations

对偶学习:探秘人工智能的对称之美 议题 (Session)

以深度学习为代表的人工智能技术通常需要大量的有标签训练数据,这对于很多应用领域而言并非易事。为了解决这个挑战,我们利用人工智能的对称之美——很多人工智能任务天然就是双向的,比如中到英翻译 vs.英到中翻译,图像分类 vs. 图像生成,语音识别 vs. 语音合成——来为机器学习建立闭环、生成有效的反馈信号,从而在缺乏有标签数据的情况下也能实现高效学习。我们将这种新型的学习方法称之为“对偶学习”。对偶学习已经被成功应用到诸多领域,取得了非同凡响的效果。本报告中,我们将针对对偶学习的数学模型、优化算法、概率解释、实验结果,收敛性分析等进行详细讨论,展示对偶学习的魅力,并对它在人工智能领域的更广泛应用进行展望。对偶学习有关的研究成果已发表在NIPS、ICML、IJCAI、AAAI等人工智能领域最顶尖的国际会议之上。

Ben Lorica is the chief data scientist at O’Reilly Media. Ben has applied business intelligence, data mining, machine learning, and statistical analysis in a variety of settings, including direct marketing, consumer and market research, targeted advertising, text mining, and financial engineering. His background includes stints with an investment management company, internet startups, and financial services.

Presentations

周五欢迎致辞 (Friday opening welcome) 主题演讲 (Keynote)

Program chairs Ben Lorica, Jason Dai, and Roger Chen open the second day of keynotes.

周四欢迎致辞(Thursday opening welcome) 主题演讲 (Keynote)

大会日程主席 Ben Lorica、Roger Chen 与 Jason Dai致辞开始第一天主题演讲。

Zhenxiao Luo is an engineering manager at Uber, where he runs the interactive analytics team. Previously, he led the development and operations of Presto at Netflix and worked on big data and Hadoop-related projects at Facebook, Cloudera, and Vertica. He holds a master’s degree from the University of Wisconsin-Madison and a bachelor’s degree from Fudan University.

Presentations

大规模人工智能在优步:大数据和机器学习的双城记 议题 (Session)

优步应用大数据技术和机器学习技术为客户寻找最舒适的出行地点,预测最佳的航行路线,从而更好的服务客户需求。在这个讲座中,我们将讨论优步如何建立起大数据系统,和机器学习系统,并逐渐将两个系统统一起来。我们会重点讨论优步大数据的缓存策略,以及如何有效的应用缓存来支持大规模的机器学习。

Sherry Moore is a software engineer on the Google Brain team. Her other projects at Google include Google Fiber and Google Ads Extractor. Previously, she spent 14 years as a systems and kernel engineer at Sun Microsystems.

Presentations

TensorFlow对科学的影响 主题演讲 (Keynote)

人工智能已经不是未来的科技,它正快速地成为我们日常生活的一部分。在本演讲中,谷歌TensorFlow的领导者Sherry Moore将会介绍机器学习是如何造福世界的,特别是对于科学的发展。她将会讨论她自己的关于学习如何学习(AutoML)的工作以及几个在中国和全世界使用TensorFlow和机器学习的迷人的案例。


在TensorFlow中构建和部署模型 议题 (Session)

TensorFlow可以让你进行高速运算,很多时候是在机器学习的情景下。 Sherry Moore将会介绍TensorFlow的最新进展,包括TensorFlow立刻执行机制和TensorFlow Lite。她还会分享一些最佳实践,并将演示机器学习的一些有用的应用。

Arsenii Mustafin is a Russian PhD student at Fudan University, where he specializes in economic studies and data analysis.

Presentations

Building deep reinforcement learning applications on BigDL and Spark 议题 (Session)

Deep reinforcement learning is a thriving area and has wide applications in industry. Arsenii Mustafin shares his experience developing deep reinforcement learning applications on BigDL and Spark.

Haikal Pribadi is the founder and CEO of GRAKN.AI, the database for AI, which uses machine reasoning to handle and interpret complex data. GRAKN.AI was recently awarded Product of the Year 2017 by the University of Cambridge Computer Lab. Haikal’s interest in the field began at the Monash Intelligent Systems Lab, where he built an open source driver for the Parallax Eddie Robot, which was then adopted by NASA. Haikal was also the youngest algorithm expert behind Quintiq’s optimization technology, that supports some of the world’s largest supply chain systems in transportation, retail, and logistics. He holds a master’s degree in AI from the University of Cambridge.

Presentations

Databases: The past, the present, and the future of cognitive computing 议题 (Session)

The relational database enabled the rise of BI systems, and NoSQL databases enabled web scale applications. Now, the future is cognitive computing. However, these systems process data that is more complex than before. Haikal Pribadi reviews the evolution of databases and explains where knowledge graphs and bases sit in this evolution. Could they serve as the next generation of databases?

Nishant Sahay is a senior architect in the Open Source COE lab at Wipro, where he is responsible for research and solution development in the area of machine learning and deep learning. Nishant has extensive experience in data analysis, design, and visualization. He has written articles on technology in online forums and presented at multiple open source conferences, such as OSI Days, GIDS, and CNCF-Kubeconf.

Presentations

Smart diagnosis in healthcare with deep learning 议题 (Session)

Deep learning with ConvNet in particular has emerged as a promising tool in medical research labs and diagnostic centers to help analyze images and scans, and systems are now surpassing human capability for manual inspection. Nishant Sahay explains how to apply deep learning to analyze high-end microscope images and X-ray scans to provide accurate diagnosis.

Kaz Sato is a staff developer advocate on the Cloud Platform team at Google, where he leads the developer advocacy team for machine-learning and data analytics products such as TensorFlow, the Vision API, and BigQuery. Kaz has been leading and supporting developer communities for Google Cloud for over seven years, is a frequent speaker at conferences, including Google I/O 2016, Hadoop Summit 2016 San Jose, Strata + Hadoop World 2016, and Google Next 2015 NYC and Tel Aviv, and has hosted FPGA meetups since 2013.

Presentations

The tensor processing unit: A processor for neural network designed by Google 议题 (Session)

The tensor processing unit (TPU) is a LSI designed by Google for neural network processing. The TPU features a large-scale systolic array matrix unit that achieves outstanding performance-per-watt ratio. Kazunori Sato explains how a minimalistic design philosophy and a tight focus on neural network inference use cases enables the high-performance neural network accelerator chip.

Zhefu Shi is a researcher at the University of Missouri, where he works on mathematical modeling, artificial intelligence, machine learning, and cloud computing. Zhefu holds a PhD in computer science, and a master’s degree in math and statistics.

Presentations

Using AI to analyze the impact of financial news 议题 (Session)

It is critical to analyze the business impact on finance market from worldwide events. Zhefu Shi explains how to use AI to analyze the impact of financial news, using a financial data pipeline. Zhefu outlines how to extract financial entity information and use it to analyze business impact. All of the components use AI to enhance functionality.

Shyam Sundar is the Sydney-based regional director of APJ at Anodot, a leading provider of AI-powered analytics. Shyam has 17 years of experience in strategic data innovation, advanced business intelligence, analytics, and anomaly detection. Previously, he worked in APJ for companies including Cloudera, HP Vertica, Sybase, and numerous startups. Shyam has also consulted on cutting-edge data strategy for Global 2000 companies. Shyam holds a bachelor of engineering from Annamalai University in India and an MBA from Monash University in Australia.

Presentations

Lessons learned from Singles Day: Using AI to keep ecommerce and internet business glitch free 议题 (Session)

Shyam Sundar explains how to use unsupervised machine learning to keep websites and mobile apps running smoothly under the stress of massive numbers such as those seen on Singles Day. With this method, pricing errors, conversion problems, and business opportunities can be caught early and resolved, protecting companies against revenue loss and brand damage.

Hendra Suryanto is chief data scientist at Rich Data Corporation. Hendra has over 20 years’ experience in data science, big data, business intelligence, and data warehousing spanning across data architecture, data science and data engineering, managing and designing end-to-end data analytics solution within Agile continuous delivery DevOps framework. Previously, Hendra was a lead data scientist in KPMG’s Advisory practice, where he advised KMPG’s clients globally in data science and big data projects, and worked for a number of leading organizations in various domain verticals, such as telecommunications, banking, fraud, risk, marketing, and insurance, including Westpac Bank, Commonwealth Bank Australia, Veda, Bupa, HCF, and Vodafone. Hendra holds a PhD in artificial intelligence, which he followed with postdoctoral research in machine learning.

Presentations

Feature engineering: The missing link in applying machine learning to deliver business value 议题 (Session)

Hendra Suryanto shares a case study from a Canadian financial lender that his company helped transition from manual to automated credit decisioning, using gradient boosting machine and deep learning to build the model. In addition to modeling techniques, Hendra highlights the role feature engineering plays in improving model performance.

David Talby is a chief technology officer at Pacific AI, helping fast-growing companies apply big data and data science techniques to solve real-world problems in healthcare, life science, and related fields. David has extensive experience in building and operating web-scale data science and business platforms, as well as building world-class, Agile, distributed teams. Previously, he was with Microsoft’s Bing Group, where he led business operations for Bing Shopping in the US and Europe. Earlier, he worked at Amazon both in Seattle and the UK, where he built and ran distributed teams that helped scale Amazon’s financial systems. David holds a PhD in computer science and master’s degrees in both computer science and business administration.

Presentations

Extending Spark NLP: Training your own deep-learned natural language understanding models 议题 (Session)

To achieve high accuracy when reasoning about text, you generally need to understand specific languages, jargons, domain-specific documents, and writing styles. David Talby explains how to train custom word embeddings, named entity recognition, and question-answering models on the NLP library for Apache Spark.

Introducing Spark NLP: State-of-the-art natural language processing at scale 议题 (Session)

Natural language processing is a key component in many data science systems that must understand or reason about text. David Talby offers an overview of the NLP library for Apache Spark, which natively extends Spark ML to provide open source, fully distributed, and optimized versions of state-of-the-art NLP algorithms, covering the library's design and sharing working code samples in PySpark.

王海峰,博士,现任百度副总裁,AI技术平台体系(AIG)总负责人,兼任百度研究院院长,负责百度研究院、语音技术部、自然语言处理部、知识图谱部、大数据部、视觉技术部、人脸技术部、增强现实技术部、AI技术生态部、及若干创新业务部门等。

王海峰博士于2010年1月加入百度。在2010-2013年期间,他先后为百度创建了自然语言处理部、互联网数据研发部(包括知识图谱和互联网数据挖掘)、推荐引擎和个性化部、多媒体部(包括语音和图像技术)、图片搜索部、语音技术部等。2013年上半年,王海峰作为执行负责人协助创建了百度深度学习研究院(IDL)。同年10月,他晋升为公司副总裁。2014年,王海峰转岗至搜索业务群组任副总经理,先后负责了百度搜索、手机百度、百度信息流、自然语言处理、百度翻译、互联网数据挖掘、知识图谱、语音搜索、图像搜索、度秘、小度机器人、百度新闻、百度手机浏览器、商业平台、糯米技术平台、Hao123等。
王海峰博士是自然语言处理领域世界上最具影响力的国际学术组织ACL(Association for Computational Linguistics)50多年历史上唯一出任过主席(President)的华人,是截至目前最年轻的ACL会士(Fellow),同时也是唯一来自中国大陆的ACL会士。王海峰还在多个国际学术组织、国际会议、国际期刊兼任各类职务。

王海峰任深度学习技术及应用国家工程实验室理事长兼主任。他同时兼任中国人工智能产业发展联盟、新一代人工智能产业技术创新战略联盟、类脑智能技术及应用国家工程实验室、中国电子学会、中国网络空间安全协会、中国中文信息学会等机构副理事长,大数据系统软件国家工程实验室技术委员会副主任,新一代人工智能战略咨询委员会委员等。

王海峰于2017年荣获首届全国创新争先奖,是唯一来自互联网行业的获奖人,于2015年获得国家科技进步奖二等奖(第一获奖人),2013、2014连续两年获得中国电子学会科学技术奖一等奖(均为第一获奖人)。
王海峰已发表学术论文百余篇,已授权或公开的专利申请120余项。王海峰曾作为负责人承担国家核高基重大专项、863重大项目,并正在承担973、自然科学基金重点项目等。

Presentations

主题演讲, Dr. Wang Haifeng 主题演讲 (Keynote)

敬请期待更多细节。

有多年的研发及架构经验,目前主要负责TalkingData移动端基于传感器数据的场景感知应用研究和开发,以及计算机视觉技术在零售行业的解决方案。毕业于北京邮电大学

Presentations

深度学习在 Android 平台的应用 议题 (Session)

目前,深度学习在移动端的应用越来越受到重视,从芯片制造商到手机厂商,一直到应用开发者,都在为在智能手机上运行深度学习模型做出了很多努,开发者一方面很难找到针对移动端优化过的解决特定应用场景的模型,一方面不知道应该如何选择这些框架,TalkingData 推出的 Android Deep Learning Framework 就为了解决这些问题。我们提供了针对移动平台的各种类型的模型,以及它们在主流机型行的实测 Benchmark,另外也提供了利用这些预训练模型和自己的数据集进行再训练的服务器端脚本和自动化工具,最后就是封装了一个上层 DL API,让开发者可以支持各种移动端深度学习框架,并为这些模型的使用提供统计分析服务。

Bichen Wu is a PhD student at UC Berkeley, where he focuses on deep learning, computer vision, and autonomous driving.

Presentations

用于无人驾驶的深度学习技术 议题 (Session)

深度学习近年来的成功极大地促进了自动驾驶技术的快速发展。但不少问题依然存在:1)深度学习模型需要大量的训练数据 2)即便是深度学习模型也很难达到100%准确率 3) 深度学习模型的计算复杂度太高,超出了车载计算机的处理能力。这个讲座将会关注以上几个问题。

Mingxi Wu is the vice president of engineering at TigerGraph, a Silicon Valley startup that is building a world-leading real-time graph data platform. Over the past 15 years, Mingxi has been focusing on database research and data management software building, serving within Microsoft’s SQL Server Group and Oracle’s Relational Database Optimizer Group. He has won research awards from the most prestigious publication venues in database and data mining (SIGMOD, KDD, and VLDB). He holds five US patents on big data and three pending international patents on graph management. He’s currently working on an easy-to-use and highly expressive graph query language. Mingxi holds a PhD from the University of Florida, where he specialized in both databases and data mining.

Presentations

为什么图模型对人工智能应用至关重要? 议题 (Session)

为了让机器像人一样思考,一个成功的人工智能应用程序的关键部分必须由强大的数据管理软件支持。在这次演讲中,我们将讨论人工智能数据管理的需求,并指出图模型的独特优势。我们将深入讨论几个现实生活中部署的,且将它们的成功归因于图模型的人工智能应用程序。

吴唯玥,牛津大学工商管理硕士。牛津大学种子基金投资经理,曾任德勤咨询汽车行业高级顾问,主要专注于销售业绩提升与企业管理流程重组优化,带领团队完成多个汽车品牌咨询项目。中国注册会计师协会会员。

Presentations

无人驾驶技术产业链条 议题 (Session)

无人驾驶技术是多个技术的集成, 一个无人驾驶系统包含了多个传感器,包括长距雷达、激光雷达、短距雷达、车载摄像头、超声波、 GPS、 陀螺仪等。每个传感器在运行时都不断产生数据,而且系统对每个传感器产生的数据都有很强的实时处理要求。 无人驾驶序幕刚启,其中有着千千万万的机会亟待发掘。在此背景之下,过去的几年中,自动驾驶产业化在多个方面取得了很大进步,其中合作共享已成为共识,产业链不断整合,业界企业相继开展合作,传感器价格将不断下降,预计在2020年,将有真正意义上的无人车面世。 我们可以预测一个不远的未来,届时所有行驶的车辆都是无人驾驶车,我们将迎来一个更加安全、更加清洁环保的世界。 本次演讲,我们将解析无人驾驶技术产业链条,分析无人驾驶发展和即将面临的问题。最后,将给出无人驾驶发展的路线图,揭示在 未来二十年内无人驾驶的走势。

加入DataVisor之前,Zhong Wu是微软Bing资深工程师,主攻多媒体搜索。工作内容广泛涵盖可伸缩特性、高绩效系统和高效算法,致力于从数十亿图像中提高搜索关联性,增强搜索索引质量。
清华大学本科、博士,博士生导师沈向洋。

Presentations

人工智能在欺诈检测中的应用 议题 (Session)

随着互联网不断发展,面向用户的线上网站服务也进入极速发展期,吸引了大量的用户,整个互联网进入“十亿用户时代”。一些有组织的欺诈团伙利用这个特点,大量创建虚假账户或盗取正常用户账户,以此潜伏在大量正常用户中,在银行、网站和手机应用软件上实施欺诈。由于规则引擎和传统机器学习模型需要经常更新、维护,而且只有在损失发生后才会生成相应反应机制,因此反欺诈团队很难领先一步走在欺诈者前面。人工智能的发展,给整个反欺诈领域带来新的机会。

Tony Xing is a senior product manager in AIDI (the AI, data, and infrastructure team) within Microsoft’s AI and research organization. Previously, he was a senior product manager on the Skype data team within Microsoft’s Application and Service Group, where he worked on products for data ingestion, real-time data analytics, and the data quality platform.

Presentations

把AI注入BI: Kensho – 微软的自动化商业指标监控和诊断工具 议题 (Session)

在这个议题中,我们会介绍Kensho, 一个基于AI的商业指标监控与诊断工具, 我们通过将AI元素注入这个BI工具,从而构建来服务不同的微软团队的历程。我们的从中学到的经验教训,技术的选择和烟花,架构,算法等等。通过工程+数据科学解决了一个工业界的一个通用需求。

Bixiong Xu is the principal dev manager on the AI, data, and infrastructure team at Microsoft.

Presentations

把AI注入BI: Kensho – 微软的自动化商业指标监控和诊断工具 议题 (Session)

在这个议题中,我们会介绍Kensho, 一个基于AI的商业指标监控与诊断工具, 我们通过将AI元素注入这个BI工具,从而构建来服务不同的微软团队的历程。我们的从中学到的经验教训,技术的选择和烟花,架构,算法等等。通过工程+数据科学解决了一个工业界的一个通用需求。

徐小磊,目前就职于新智新氦科技有限公司,担任算法工程师。新氦科技是新智集团下属,上海的一家大数据基础架构公司。徐小磊目前主要负责新氦深度学习云平台的搭建和基于深度学习,深度强化学习的自然语言处理应用研发工作。

Presentations

基于TensorFlow的高效交互式深度学习平台及应用(An efficient and interactive deep learning platform with TensorFlow) 议题 (Session)

目前单机多卡训练是深度学习的标配,但是单机的GPU数目总有上限,因此如何通过多机多卡进行高效的分布式训练就尤其重要。比如,如何将简单的单机程序快速部署到多机并得到相应的加速比,如何使得对GPU的调度与大数据处理平台无缝对接,并使GPU成为平台上按需调度、动态扩容的资源,这些问题的解决对算法迭代优化起到关键作用。 本次talk会详细介绍如何基于Kubernetes和Docker构建TensorFlow的微服务化应用,具体从以下几个方面展开:从少量样本数据的单机快速原型设计验证,无缝切换到大量全数据的多机多卡分布式训练过程;一键启动分布式训练,即基于新氦定制的深度学习云平台,用户无需关注分布式细节,可直接通过可视化web界面进行分布式参数配置和训练代码提交,并可实时可视化监控模型训练收敛性、系统资源消耗和模型输出日志等;模型训练结束后可实时serving将模型快速部署到生产环境。

Yu Xu is the founder and CEO of TigerGraph, the world’s first native parallel graph database. He is an expert in big data and parallel database systems and has over 26 patents in parallel data management and optimization. Previously, Yu worked on Twitter’s data infrastructure for massive data analytics and was Teradata’s Hadoop architect leading the company’s big data initiatives. Yu holds a PhD in computer science and engineering from the University of California, San Diego.

Presentations

为什么图模型对人工智能应用至关重要? 议题 (Session)

为了让机器像人一样思考,一个成功的人工智能应用程序的关键部分必须由强大的数据管理软件支持。在这次演讲中,我们将讨论人工智能数据管理的需求,并指出图模型的独特优势。我们将深入讨论几个现实生活中部署的,且将它们的成功归因于图模型的人工智能应用程序。

Hua Yang is a director of eBay’s search team in Shanghai, where he is part of eBay’s global search product team. He and his team are researching and developing machine learning and artificial intelligence technologies to provide the most relevant search experiences for eBay customers. Hua has 20 years of software development experience, including over a decade in digital advertising and search engine development at both Microsoft and eBay. Hua studied software engineering and artificial intelligence and holds a PhD in computer science from Vanderbilt University.

Presentations

AI technologies on eBay's search platform 议题 (Session)

The search engine has been a great platform for machine learning technologies, and the latest developments in AI open a new frontier, transforming the search engine into an AI platform. Hua Yang explores the deep learning and natural language understanding technologies used in eBay's ecommerce search platform.

Season Yang is a data scientist in residence at the Data Incubator, where he also contributes to curriculum development and instruction. Previously, Season worked at NASA’s Goddard space center, where he studied climate change models with data analysis. He holds a double bachelor’s degree in applied mathematics and scientific computation and economics from UC Davis and a master’s in applied mathematics from Columbia, specializing in numerical computation.

Presentations

用TensorFlow进行深度学习 2天培训 (2-day Training)

TensorFlow是一个流行的深度学习的工具。我们会介绍TensorFlow的流程图、学习使用它的Python API,并展示它的用处。我们会从简单的机器学习算法开始,然后实现神经网络。我们还会讨论一些真实的深度学习的应用,包括机器视觉、文本处理和生成型网络。

PerceptIn创始工程师以及技术负责人,清华大学微电子所博士。目前在PerceptIn负责无人驾驶产品。

Presentations

PerceptIn低成本无人驾驶解决方案 议题 (Session)

得益于人工智能和机器人技术的快速发展,无人驾驶技术逐渐成熟,预计将会孕育出一个万亿规模的市场,并深刻地改变人们的交通出行方式。我们认为低速限制性的园区将会首先大规模部署无人驾驶技术,首先因为限制性园区对无人驾驶应用的需求巨大,其次由于驾驶环境简单限制性园区容易实现无人驾驶,第三从成本角度考虑,大规模部署无人驾驶方案成本需要在万元美金以内。所以,这里我们将主要探讨适用于限制性园区的低成本无人驾驶解决方案。

Reza Bosagh Zadeh is founder and CEO at Matroid and an adjunct professor at Stanford University, where he teaches two PhD-level classes: Distributed Algorithms and Optimization and Discrete Mathematics and Algorithms. His work focuses on machine learning, distributed computing, and discrete applied mathematics. His awards include a KDD best paper award and the Gene Golub Outstanding Thesis Award. Reza has served on the technical advisory boards of Microsoft and Databricks. He is the initial creator of the linear algebra package in Apache Spark. Through Apache Spark, Reza’s work has been incorporated into industrial and academic cluster computing environments. Reza holds a PhD in computational mathematics from Stanford, where he worked under the supervision of Gunnar Carlsson. As part of his research, Reza built the machine learning algorithms behind Twitter’s who-to-follow system, the first product to use machine learning at Twitter.

Presentations

Scaling convolutional neural networks with Kubernetes and TensorFlow 议题 (Session)

Reza Zadeh offers an overview of Matroid’s Kubernetes deployment, which provides customized computer vision and stream monitoring to a large number of users, and demonstrates how to customize computer vision neural network models in the browser. Along the way, Reza explains how Matroid builds, trains, and visualizes TensorFlow models, which are provided at scale to monitor video streams.

Turning machine learning research into products for industry 主题演讲 (Keynote)

Reza Zadeh details three challenges on the way to building cutting-edge ML products, with a focus on computer vision, offering examples, recommendations, and lessons learned.

Greg Zaharchuk is an associate professor in radiology at Stanford University and a neuroradiologist at Stanford Hospital. His research interests include deep learning applications in neuroimaging, imaging of cerebral hemodynamics with MRI and CT, noninvasive oxygenation measurement with MRI, clinical imaging of cerebrovascular disease, imaging of cervical artery dissection, MR/PET in neuroradiology, and resting-state fMRI for perfusion imaging and stroke.

Presentations

深度学习与人工智能在神经影像中的前沿应用 议题 (Session)

人工智能与深度学习正在快速改变医疗产业发展。本讲座将介绍斯坦福的深度学习领域学者与斯坦福医院医生、教授合作研发的技术,以及如何快速地优化临床医学影像的使用。人工智能技术让医学影像的采集与处理更加快速、高效、便捷与智能。 具体技术应用包括: 1. 如何通过人工智能优化临床影像流程,优化诊断治疗规划 2. 如何通过人工智能与深度学习预测神经疾病病人的预后和疾病发展 3. 如何通过人工智能与深度学习技术加速神经影像流程 4. 如何通过人工智能与深度学习技术显著减少放射性与显影剂使用

Ruiwen Zhang is a senior research statistician at SAS, focusing on machine learning and data mining. She holds a PhD from the Department of Statistics and Operation Research at the University of North Carolina at Chapel Hill.

Presentations

Representing knowledge through graphical models 议题 (Session)

Drawing on several real-world cases, Ruiwen Zhang demonstrates how to visualize the structure of a probabilistic model and provide better insights into the model properties, which can be further used to design and motivate new models, and how to reduce the computational complexity required to perform inference and learning in sophisticated models using graphical models.

Xiatian Zhang is chief data scientist at TalkingData, where he is responsible for mobile big data mining and ML algorithm research and implementation. Xiatian has long engaged in data mining and machinelearning research and has dozens of research papers in publication and sufficient patents. Previously, he worked for IBM’s China Research Institute, the Tencent data platform, and Huawei’s Noah’s Ark Lab.

Presentations

Smart Data – 从数据驱动智能到智能驾驭数据 议题 (Session)

大数据直接推动了人工智能的发展,但如何有效管理和利用大数据也一直是非常有挑战的问题。梳理数据,整理数据,利用数据都非常依赖于数据工程师,数据分析师和数据科学家的个人能力,经验,以及责任心。基于数据创造和发展智能的一大瓶颈就在于这个过程非常的依赖于人。为了提高效率,降低基于数据的智能的成本,扩大其应用范围,我们必须利用智能技术来处理和利用大数据,尽量减少对人的依赖。

Yi Zhang is a tenured professor at the University of California, Santa Cruz, and cofounder and CTO of Rulai. Yi has 20 years of research experience in AI. She has received various awards, including an ACM SIGIR best paper award, the National Science Foundation Faculty Career Award, a Google research award, a Microsoft research award, and an IBM research fellowship. She has been a program cochair, area chair, and PC member for various top international conferences. Yi has served as a consultant or a technical adviser for several companies and startups, including Alibaba, Toyota, and HP. She holds a PhD in computer science from Carnegie Mellon University.

Presentations

Chatbots: How business leaders can avoid pitfalls and control AI 议题 (Session)

Yi Zhang offers a comprehensive overview of the technology landscape of the chatbot. You’ll learn best practices with regard to evaluating technologies, how to assemble the right team to manage the process, user-centered bot design principles, and risk management. Along the way, Yi share bot use cases within several industries.

周明博士,微软亚洲研究院副院长、国际计算语言学协会(ACL)候任主席、中国计算机学会理事、中文信息技术专委会主任、术语工作委员会主任、中国中文信息学会常务理事、哈尔滨工业大学、天津大学、南开大学、山东大学、北航等多所学校博士导师,微软-清华大学联合实验室主任。周明博士1985年毕业于重庆大学,1991年获哈尔滨工业大学博士学位。1991-1993年清华大学博士后,随后留校任副教授。1996-1999访问日本高电社公司领导中日机器翻译研究。他是中国第一个中英翻译系统CEMT-I(哈工大1989年)、日本最有名的中日机器翻译产品J-北京(日本高电社1998年)的研制者。1999年,周明博士加入微软亚洲研究院,不久开始负责自然语言研究组。他带领团队进行了微软输入法、英库词典(必应词典)、中英翻译、微软中国文化系列(微软对联、微软字谜、微软绝句)等重要产品和项目的研发,并对微软Office、必应搜索、Windows等产品中的自然语言技术做出了重要贡献。近年来,周明博士领导研究团队与微软产品组合作开发了微软小冰(中国)、Rinna(日本)、Zo(美国)等聊天机器人系统。周明博士发表了120余篇重要会议和期刊论文(包括50篇以上的ACL文章),拥有国际发明专利40余项。他多年来通过微软与中国和亚太地区的高校合作计划,包括微软-高校联合实验室、微软实习生计划、微软-高校联合培养博士生计划、青年教师铸星培养计划,与高校和学术组织联合举办暑期学校和学术会议等多种形式,对推动自然语言处理在中国和亚太的卓越发展做出了杰出贡献。

Presentations

计算机创作对联、诗歌和音乐 议题 (Session)

创作诗歌、音乐是人类独具的能力。然而,随着深度神经网络和大数据的发展,计算机已经逐步具备了创作诗歌和音乐的能力。我们致力于把AI融入到创作过程中,并且帮助普通实现创作梦想。为此,我们长期以来进行了对联、诗词的研究。2005年就开发了中文对联系统(http://duilian.msra.cn).。以后又陆续开发了格律诗写作,猜字谜和出字谜。2016年开发了小冰写诗。目前我们正在探索先进的神经网络和大数据来模仿人类的音乐创作过程。我们采用了融入上下文的编码-解码方法来产生诗歌、歌词和谱曲。取得了富有希望的成果。我们的电脑音乐创作已经在CCTV的机智过人节目播出。获得好评,由电脑写出歌词,然后配上曲谱,然后通过声音合成,唱出歌曲。

Li Zhou is the global chief architect and principal software manager of Microsoft’s Xiaoice project, an artificial intelligent chatbot based on emotion computing. Previously, he worked on Windows Antimalware Engine, Bing China, and Cortana China. Li holds a PhD from the University of Southern California and a BS from Peking University in computer science.

Presentations

Xiaoice: Lessons learned from conversations between humans and AI 议题 (Session)

Since its first release in May of 2014, more than 100 million users in China, Japan, and the US have interacted with renowned AI product Xiaoice (小冰), which builds human-like conversation. Li Zhou shares key lessons learned from the past four years and explains how to use them to build a better chatbot experience.

转转公司架构算法部负责人,前58集团技术委员会主席,资深架构师,“架构之美”公众号作者。擅长系统架构设计,大数据,机器学习等技术领域。代表公司多次参加 QCon、ArchSummit、SDCC、CCTC、DTCC、Top100、Strata + Hadoop World、WOT 等大会嘉宾演讲,并为《程序员》杂志撰稿 2 篇。 前百度高级工程师,参与百度社区搜索部多个基础系统的设计与实现。毕业于浙江大学。

Presentations

人工智能时代,二手交易平台的智能推荐系统如何演进 议题 (Session)

转转的推荐系统从0开始打造,针对业务的不同阶段,一步步发展演进。在发展的过程中经历了全局无个性化推荐阶段、个性化离线推荐阶段、个性化实时推荐阶段、机器学习排序推荐阶段等。 本文会详细讲解不同发展阶段的原因、架构&算法的演进,让同学们对二手交易平台的智能推荐系统能够深刻认识。

10年+的数据工程经验。在携程担任过传统BI的分析经理。在顺丰担任过大数据部总监。目前在华为担任人工智能架构主任工程师。

Presentations

华为人工智能平台的探索与实践 议题 (Session)

(1)阐述下华为是怎么定义与认识人工智能这座山峰的 (2)华为的人工智能平台逻辑架构 (3)华为的这座人工智能山峰,在自然语言处理与机器学习中的技术栈 (4)在NLP+ML上的应用探索 (5)构建细而窄领域的知识图谱的探索及应用

徐晓任职于AliOS-推荐算法引擎组。从业经验包括AliOS 应用商店/信息流推荐系统的开发和优化;AliOS算法中台研发;RecSys Challenge 2016优胜者。

Presentations

Spark+BigDL 基于Hadoop的推荐系统的深度学习实践 议题 (Session)

随着深度学习的发展,其在推荐领域的可能性也被不断拓展,越来越多的基于深度学习的推荐算法在学术论文中被提出,比如:Google提出的Wide&Deep网络结构。 目前,很多大型推荐系统均构建在Hadoop生态上,而主流的深度学习工具(如:TensorFlow/Caffe/Torch)则更适合于gpu集群。因此,运行在Spark环境上的BigDL是非常合适于推荐系统的深度学习解决方案。 本议题将通过案例的形式,分享使用Spark与BigDL构建深度神经网络来优化现有推荐系统的经验。本议题的主要关注点是:如何在推荐工程中高效而健壮的实施深度学习,包括:技术选型的思考,实验场景的搭建,神经网络配置脚本的定制,模型数据的IO,自定义神经网络组件的开发等。

Grace Lee is the author of Technical Analysis and Practice in TensorFlow and the founder of the TensorFlow communication community. She is active in the technical community and is known for her answers to programming questions. Grace has deep experience with TensorFlow, source code analysis and application in different areas, processing images, social text data emotional analysis, and data mining. She participated in the the autopilot two-dimensional perception system hackathon competition based on deep learning. Previously, she was a deep learning engineer at Baidu.

Presentations

TensorFlow下的构建高性能神经网络模型的最佳实践 议题 (Session)

随着神经网络算法在图像、语音等领域都大幅度超越传统算法,但在应用到实际项目中却面临两个问题:计算量巨大及模型体积过大,不利于移动端和嵌入式的场景;模型内存占用过大,导致功耗和电量消耗过高。因此,如何对神经网络模型进行优化,使尽可能不损失精度的情况下,能减少模型的体积,并且计算量也降低,就是我们将深度学习在更广泛地场景下应用时要解决的问题。本次讲解主要着眼于在安防、工业物联网、智能机器人等设备,需要解决图像、语音场景下深度学习的加速问题,减小模型大小及计算量,构建高性能神经网络模型。

PerceptIn中国商务副总裁
毕业于清华大学自动化系,多年从事于人工智能以及计算机视觉方向的研发,产品,商务方面工作,拥有丰富经验

Presentations

视觉智能及其在机器人行业中的应用 议题 (Session)

本演讲主要阐述视觉智能(Visual Intelligence)的定义,传感器分类和介绍,流行算法和介绍,应用场景以及创新点。 介绍视觉传感器的发展历史以及分类,包括被动光摄像头和主动光摄像头以及其他衍生传感器 介绍基于视觉的算法:深度学习算法和SLAM算法 介绍视觉智能在机器人行业中的应用,包括家庭机器人,服务类机器人,无人驾驶汽车。 最后介绍多传感器融合的解决方案在机器人行业的应用以及必要性。

自2012年Hinton教授等人使用基于卷积神经网络的深度学习模型赢得了ImageNet分类比赛以来,深度学习的热潮席卷了各个行业。文章在介绍了深度学习历史的基础上,探索了国内地质行业中,深度学习模型的使用情况。并介绍了深度学习的基础概念,如神经元,神经网络以及监督学习和无监督学习。在概念的基础上,介绍了深度学习基础模型中的两个重要的网络,深度置信模型(DBN)和卷积神经网络(CNN)。文章的最后,类比深度学习在医学中的应用情况,提出了深度学习在地质学中潜在的应用前景。

Presentations

深度学习与地质学能碰撞出什么样的火花? 议题 (Session)

众所周知,现在的深度学习已经在各个行业开始了应用。但是深度学习如何与地质行业相结合,这还是一个新兴的话题,国外目前,已经开始用深度学习来处理实验室地震数据,用以提高地震预测的时间;国内也已经有很多人用卷积神经网络开始对岩石图像数据进行处理,这次议题我做的报告是,在介绍前人工作的基础上,介绍一下自己在地质上的应用!

目前在阿里计算平台事业部大规模算法团队负责大规模深度学习算法基础设施相关建设工作,对大规模分布式机器学习的开发、建设、优化以及在不同业务场景中的落地应用有较为深入的理解和认识。之前先后在奇虎360担当广告技术部门架构师,Yahoo北京研发中心担当效果广告系统技术负责人。

Presentations

端到端深度学习优化在互联网业务场景下的应用实践 议题 (Session)

本议题会分享我们在典型互联网业务场景(图像、文本处理等)下的深度学习优化实践经验,包括离线训练和在线Inference,并会从系统与算法相结合的角度进行相关经验的阐述和介绍。

焦加麟,互联网及移动互联网资深技术从业人员,专注LBS(Location-Based Service)、高清地图、大数据、机器学习、人工智能、搜索等领域的研发,曾任职于微软搜索引擎Bing、IBM T.J Watson Research,现在Uber美国总部从事地图相关工作。具有中山大学计算机科学本科、上海交通大学计算机科学硕士和美国密西根大学电子工程硕士学位。 Linkedin Profile:https://www.linkedin.com/in/jialin-jiao-b3852711/

Presentations

人工智能在高精地图制作中的应用 议题 (Session)

在无人车科学家和工程师们孜孜不倦的实践和思辨中,高精地图(High Definition Map)事实上已经成为现今无人车技术生态系统中的不可缺少的基础设施之一。同样是对现实世界道路网络以及周边环境的建模,比起一般的电子地图,高精地图必须精确到厘米级,同时需要更频繁的更新以保证其正确性。如此高度的精确性和频繁更新的要求,给高精地图的制作带来来巨大的挑战,其中包括专用软硬件的设计和研发、成千上万的城市的天文数字级别的数据的收集、处理、存储和信息化、语义化等等。这一切,使得高精地图的制作成本非常昂贵,需要耗费大量的时间和人力。利用人工智能提高自动化的程度,是降低成本、加快其制作过程的必须的手段。本议题将会深入浅出的介绍各种人工智能技术在高精地图的制作中的各个环节中的应用,以科普大众并唤起专业人士对人工智能在无人车高精地图制作中的应用的兴趣和重视。

王刚是小米小爱团队的负责人,为小米的各类智能设备提供语音交互的服务,目前已支持小米人工智能电视、小米AI音箱及生态链其他智能硬件。王刚毕业于香港科技大学计算机科学系博士,长期从事数据挖掘和机器学习方向的研究,2007年加入微软亚洲研究院,2010年加入腾讯,2013年加入小米。

Presentations

小米语音交互的最新进展、面临的难题以及展望 议题 (Session)

本次讲演将分享小米语音交互在产品和技术方面的最新进展和面临的一些难题,以及对未来语音技术发展的展望。

韩建军,1972年7月30日生,现为华中科技大学计算机科学与技术学院副教授、硕士生\博士生导师,中国计算机学会嵌入式系统专委会委员。曾以访问学者身份在美国UC Irvine以及韩国首尔大学等地学术交流。主要研究领域: 实时系统、调度算法、高性能计算。荣获华中科技大学2005年优秀博士论文奖。

主持国家自然科学基金面上(3项)、中国博士后科学基金(一等)、教育部高校基金、航天科工集团示范项目、华为公司研究基金, 作为核心成员承担国家自然科学基金重点、科技部(863、973)等项目。在TC、TPDS、TCAD、TECS、TSUSC、软件学报、计算机研究与发展、通信学报等国内外期刊和IEEE会议(IPDPS、ICPP)上发表学术论文50余篇,被SCI、EI收录近40篇。其中第一作者发表CCF推荐A/B类论文10余篇。

现为TC、TPDS、TECS、TSUSC、JSA、IEEE Systems Journal、FGCS、JCST、电子学报、RTSS、RTAS、IPDPS、DAC、ICPP等期刊及会议的审稿人/PC member。已独立培养硕士毕业20余人,培养博士毕业5人。承担《算法设计与分析》、《并行计算》以及《程序设计方法学》等硕士生及本科生教学。

Presentations

多核嵌入式智能系统的实时调度策略及实现 议题 (Session)

嵌入式AI与云端AI的协同融合已成为当今人工智能计算系统的主流方式。首先介绍嵌入式AI的应用范围、特点及其发展趋势。面向异构多核+特定加速器的嵌入式计算系统,基于资源共享的多核体系结构,结合无人驾驶、机器人等AI领域的混合关键实时系统,针对制约实时应用效率提升的关键因素,从实时应用的调度算法、调度策略及Linux操作系统实现等方面,汇报当前的研究进展。侧重多核系统中资源竞争限制下的实时可调度理论、划分调度算法、节能调度机制、操纵系统实现等相关内容,介绍目前的高效调度策略及技术实现方案,用以提高嵌入式智能系统的资源利用率、并行效能及能量效率。 面向嵌入式AI系统的发展趋势,从主流的计算平台体系结构的特征分析出发,提出当前实时调度理论及应用实现中尚存的关键问题,共同探讨可行解决方案及技术手段,为奠定嵌入式AI系统中实时应用的理论及实践的基础提供有益思路。

黄铃,慧安金科(北京)科技有限公司创始人、CEO,清华大学交叉信息研究院兼职教授。主要技术背景是人工智能、信息安全和金融风控。他是全球为数不多的同时精通人工智能和计算机安全的顶级专家,在美国加州大学伯克利分校获得计算机科学博士 (2002-2007),师从 Anthony Joseph 和 Michael Jordan ,从事机器学习算法研究以及计算机网络建模应用。他是美国硅谷著名的反欺诈公司DataVisor的创始成员和大数据总监 (2014-1016),主持了公司整个机器学习,用户行为分析和信用分析系统。他在美国英特尔研究院任资深科学家七年(2007-2014),和 Intel McAfee 开展多个合作项目,应用人工智能技术解决网络和数据安全问题。他在人工智能,大数据分析和金融科技相关领域有近十五年的研究和开发背景,在世界顶尖会议上发表近50篇论文,在 Google Scholar 上总引用已超过5,000次。

Presentations

人工智能和金融科技:量化金融信用与欺诈风险的评估 教学辅导课 (Tutorial)

您想了解金融企业是怎样利用大数据和人工智能技术来画像个人行为并检测欺诈用户的吗?互联网金融幕后的量化分析流程是怎么杨的?个人信用是怎样通过大数据被量化的?在实践过程中,机器学习算法的应用存在着哪些需要关注的方面?怎样通过图谱分析来融合多维数据,为我们区分正常用户和欺诈用户? 这套辅导课基于清华大学交叉信息研究院2017年春天新开设的一门"量化金融信用与风控分析”研究生课。其中会用LendingClub的真实借贷数据做为案例,解说一些具体模型的实现。