Schedule List ViewGrid View

Q&A with Kai Fu Lee

Magnus Hyttsten and Priya Gupta demonstrate how to perform distributed TensorFlow training using the Keras high-level APIs. They walk you through TensorFlow's distributed architecture, how to set up a distributed cluster using Kubeflow and Kubernetes, and how to distribute models created in Keras.

In spite of the enormous excitement about the potential of deep learning, several key challenges—from prohibitive hardware requirements to immature software offerings—are impeding its widespread enterprise adoption. Evan Sparks and Ameet Talwalkar detail fundamental challenges facing organizations looking to adopt deep learning and present novel solutions to overcome several of them.

Woj Zaremba discusses deep reinforcement learning for robotics.

Earlier this year, Amazon, Facebook, and Microsoft partnered to create the Open Neural Network Exchange (ONNX)—an open format to represent deep learning models. Sarah Bird explains in detail how the ONNX framework can help you take AI from research to reality as quickly as possible.

Join Ari Kaplan, Erin LeDell, Chris Coad, and David Kearns to see where AI meets business intelligence, as they explore the latest ML technologies and concepts powering today's decisions, including Hortonworks, Aginity Amp, H2O.ai, IBM Data Science Experience, and more—using real-life baseball data to illustrate the concepts.

In neuroscience, the wiring diagram of the brain is a connectome. Jana Eggers and Elsie Kenyon built a connectome of the AI/ML "brain" via arXiv papers. They share the results of how papers, topics, keywords, authors, institutions, publication dates, citations, and more are linked and with what strength, offering interesting insights on how the AI research world is connected.

Ira Cohen shares a novel approach for building more reliable prediction models by integrating anomalies in them. Arun Kejariwal then walks you through how to marry correlation analysis with anomaly detection, discusses how the topics are intertwined, and details the challenges you may encounter based on production data.

KC Tung explains why LSTM provides great flexibility to model the consumer touchpoint sequence problem in a way that allows just-in-time insights about an advertising campaign's effectiveness across all touchpoints (channels), empowering advertisers to evaluate, adjust, or reallocate resources or investments in order to maximize campaign effectiveness.

Jian Wu discusses an end-to-end engineering project to train and evaluate deep Q-learning models for targeting sequential marketing campaigns using the 10-fold cross-validation method. Jian also explains how to evaluate trained DQN models with neural network-based baseline models and shows that trained deep Q-learning models generally produce better-optimized long-term rewards.

Danny Goodman discusses reinforcement learning and the future of software.

What does it mean to explain a machine learning model, and why is it important? Armen Donigian addresses those questions while discussing several modern explainability methods, including traditional feature contributions, LIME, and DeepLift. Each of these techniques offers a different perspective, and their clever application can reveal new insights and solve business requirements.

Drawing on the McKinsey Global Institute's groundbreaking research, Mehdi Miremadi explores commonly asked questions relating to AI and its impact on work, including: How big is the AI opportunity? Which sectors and functions will capture the most value? As AI takes hold, will there be enough jobs for humans? What will they be, and what skills are needed? What are the implications for leaders?

Ashok Srivastava explains how to make your organization AI ready, determine the right AI applications for your business and products, and accelerate your AI efforts with speed and scale.

New AI solutions in question answering, chatbots, structured data extraction, text generation, and inference all require deep understanding of the nuances of human language. David Talby shares challenges, risks, and best practices for building NLU-based systems, drawing on examples and case studies from products and services built by Fortune 500 companies and startups over the past six years.

Now more than ever, success in AI requires expertise in multiple disciplines, including big data, efficient software, and novel models and algorithms. Forrest Iandola shares an approach for developing a full stack AI team that can use all of these diverse skills to execute on industrial-scale AI problems.

Data-driven companies making intelligent products must design for security and privacy to be competitive globally. Amanda Casari details the high-level changes that EU General Data Protection Regulation (GDPR)-compliant businesses face and how this translates to teams designing products driven by machine learning and artificial intelligence.

Allison Duettmann offers an overview of AI philosophy and explains why traditional approaches need updating, because they pave the way for a singleton AI that will hardly be benevolent. Allison then discusses potential alternative AI safety strategies and their shortcomings and shares a brief survey of interesting problems in AI safety and what we can hope for if we get it right.

Technology has opened up access to more information than ever before, but it’s still on humans to turn that data into knowledge. To solve this problem, organizations are turning to AI and natural language processing to augment human analysts. Sean Gourley explores how the world’s largest organizations use AI to summarize thousands of documents and scale human analysts.

Danny Lange discusses the role of intelligence in biological evolution and learning and demonstrates why a game engine is the perfect virtual biodome for AI’s evolution. You'll discover how the scale and speed of simulations is changing the game of AI while learning about new developments in reinforcement learning.

After a slow start, the finance industry is quickly catching up with others in its adoption of AI. Joe Rothermich explains how Thomson Reuters Labs is using AI to perform research in building quantitative investment models and discusses the company's research in deep learning for credit risk, machine learning and NLP for unlocking alternative datasets, and financial graph-based analytics.

Given the revolution in data and healthcare, Shelley Zhuang predicts how certain sectors may unfold and shares opportunities for innovation. Along the way, Shelley discusses innovations that advance precision medicine by bringing together interdisciplinary fields across biology, engineering, data science, and clinical care.

Rachael Rekart offers an overview of Autodesk Virtual Agent (AVA), which has revolutionized the way Autodesk approaches customer service. Customers chat with AVA as they would a human, in natural language, and AVA processes transactions quickly, returns accurate answers, or gathers information to pass to a human counterpart to resolve the query.

The future isn’t about AI or machine learning; it’s about human-machine teaming. Candace Worley explains why, as long as there are human adversaries behind cybercrime and cyberwarfare, there will always be a critical need for the human beings teamed with machines in cybersecurity.

TensorFlow is one of the world’s biggest open source projects, and it continues to grow in adoption and functionality. Laurence Moroney, Edd Wilder-James, and Sandeep Gupta share major recent developments and highlight some future directions. Join in to learn how you can become more involved in the TensorFlow community.

Laurence Moroney dives into machine learning, AI, deep learning, and more and explains where they fit in the programmers toolkit. Along the way, he walks you through what it's all about, cutting through the hype to show the opportunities that are available in machine learning.

TensorFlow.js is the recently released JavaScript version of TensorFlow that runs in the browser and Node.js. Nick Kreeger and Ping Yu offer an overview of the TensorFlow.js ML framework and demonstrate how to perform the complete machine learning workflow, including training, client-side deployment, and transfer learning.

Richard Wei and Andrew Selle discuss Swift for TensorFlow and TensorFlow Lite, covering the current status of development and the latest developments. They then teach you how to prepare your model for mobile and how to write code that executes it on a variety of different platforms.

As machine learning evolves from experimentation to serving production workloads, so does the need to effectively manage the end-to-end training and production workflow including model management, versioning, and serving. Clemens Mewald offers an overview of TensorFlow Extended (TFX), the end-to-end machine learning platform for TensorFlow that powers products across all of Alphabet.

Michelle Casbon offers an overview of Kubeflow. By providing a platform that reduces variability between services and environments, Kubeflow enables applications that are more robust and resilient, resulting in less downtime, quality issues, and customer impact. It also supports the use of specialized hardware such as GPUs, which can reduce operational costs and improve model performance.

Neta Zmora offers an overview of Distiller, an open source Python package for neural network compression research. Neta discusses the motivation for compressing DNNs, outlines compression approaches, and explores Distiller's design and tools, supported algorithms, and code and documentation. Neta concludes with an example implementation of a compression research paper.

One of the biggest challenges in AI is how to translate advances in the lab into large-scale applications. Casimir Wierzynski reviews current trends in the field and shares case studies to illustrate why codesigning these components in concert will be critical for building the AI systems of the future.

Here at the dawn of the cognitive era, we must focus our design talents upon a new type relationship with machines—machines that can hold a conversation, interpret nonverbal cues, and draw from vast stores of human knowledge. But what should these relationships look like? Adam Cutler explains why designing for AI requires new considerations and new rules.

We now live in the age of assistive and AI companions, where everything is coming alive. Danielle Krettek demonstrates how to design for the "invisible" emotional layer of experience that marks this new wave of technology.

The public dialogue around artificial intelligence takes place at extremes, with some focusing on how robots will take everyone’s job and others speaking of a utopia where AI will do everything from cure cancer to solve world hunger. Irmak Sirer offers an alternative to these extreme views of AI: instead of artificial intelligence, we instead must focus on augmenting people’s intelligence.

Join in for a panel discussion on how AI and data are transforming the enterprise as we know it. Panelists will discuss data, talent, and technology challenges, enterprise adoption barriers, areas of disruption, and talent concerns (enterprises' willingness to pay) as AI turns the enterprise world as we know inside and out.

Transfer learning enables you to use pretrained deep neural networks and adapt them for various deep learning tasks (e.g., image classification, question answering, and more). Join Wee Hyong Tok and Danielle Dean to learn the secrets of transfer learning and discover how to customize these pretrained models for your own use cases.

Uber has implemented an ML and NLP system that suggests the most likely solutions to a ticket to its customer support representatives, making them faster and more accurate while providing a better user experience. Piero Molino, Huaixiu Zheng, and Yi-Chia Wang describe how Uber built the system with traditional and deep learning models and share the lessons learned along the way.

Deep learning is often called a black box, so how do you diagnose and fix problems in a deep neural network (DNN)? Avesh Singh and Kevin Wu explain how they systematically debugged DeepHeart, a DNN that detects cardiovascular disease from heart rate data. You'll leave with an arsenal of tools for debugging DNNs, including Jacobian analysis, TensorBoard, and DNN unit tests.

Modern radiological lung cancer screening is an entirely manual process, leading to high costs and inter-reader variability. Daniel Golden offers an overview of a deep learning-based system that automatically detects and segments lung nodules in lung CT exams and explains how it was tested for safety and efficacy. The system is FDA cleared and segments nodules as accurately as a clinician.

Complex diseases like Alzheimer’s cannot be cured by pharmaceutical or genetic sciences alone, and current treatments and therapies lead to mixed successes. Ayin Vala explains how to use the power of big data and AI to treat challenging diseases with personalized medicine, which takes into account individual variability in medicine intake, lifestyle, and genetic factors for each patient.

Captricity has deployed a machine learning pipeline that can read handwriting at human-level accuracy. Ramesh Sridharan discusses the big ideas the company learned building and deploying this system, using data to identify specific problems to solve using AI and to evaluate and validate the algorithm itself and the overall system once deployed.

Simon Crosby details an architecture for learning on time series data using edge devices, based on the distributed actor model. This approach flies in the face of the traditional wisdom of cloud-based, big-data solutions to ML problems. You'll see that there are more than enough resources at “the edge” to cost-effectively analyze, learn from, and predict from streaming data on the fly.

Over the last few years, convolutional neural networks (CNN) have risen in popularity, especially for computer vision. Anirudh Koul explains how to bring the power of convolutional neural networks and deep learning to memory- and power-constrained devices like smartphones.

Shaoshan Liu explains how PerceptIn built a reliable autonomous vehicle with a total cost under $10,000.

Alex Wong discusses some of the operational challenges associated with scalability and explainability in deep learning for real-world operational scenarios and explains how these are being tackled to enable more seamless, accessible, deployable, and transparent deep learning design and development through advancements made in the respective areas.

Using novel techniques in model-free deep reinforcement learning and control theory, Cathy Wu explores and quantifies the potential impact of a small fraction of automated vehicles on low-level traffic flow dynamics, such as congestion on a variety of important traffic contexts.

Data is your most important asset, so you don't want to move it around too much just to cater to varying compute needs. Machine learning (ML) is no exception to this rule. Ujjwal Ratan shares an end-to-end workflow for building an ML pipeline.

Artificial intelligence requires a low-latency, high-throughput storage system to keep the compute layer fully saturated with data. Liran Zvibel demonstrates why NVMe-optimized, distributed filesystems are ideal storage solutions to support AI applications and introduces a next-gen massively parallel shared filesystem that's NAND flash and NVMe optimized, built to solve the I/O starvation problem.

The time to gain market advantage through AI is now. By 2021, Gartner projects that 40% of new enterprise applications will include AI, but the majority of AI-enhanced use cases have not been developed. Ranjeeta Singh and Nick Switanek shed light on pain points and use cases where AI can drive business value for companies that make the right investments in people, process, and technology.

Ben Vigoda discusses idea learning, a new approach to deep learning. Idea learning can turn the customer journey into a personalized natural language conversation using deep, probabilistic, Turing-complete models that learn more from business people providing ideas and much less from supervised (labeled) data.

David Martinez discusses an AI canonical architecture suitable for a number of different classes of applications and shares examples focused on cybersecurity to illustrate an application area that benefits from an end-to-end AI architecture.

Dmitry Rizshkov offers an overview of OpenVINO and explores real customer case studies that exceeded the most challenging inference requirements.

Do you have constantly changing business environments across many business units and processes with multiple job schedulers and infrastructure platforms? Do you struggle with end-to-end visibility and a lot of alerts? Award-winning ignio can help. Jayanti Murty explains how and shares real-world examples of companies that have reduced operational risks and outages and technology and labor costs.

As AI grows its reach throughout society and makes decisions that affect people, any business looking to capitalize on AI’s potential must also acknowledge its impact. Kishore Durg and Teresa Escrig explain why businesses must teach and test their AI systems to act as responsible representatives so that they reflect business and societal norms of responsibility, fairness, and transparency.

What if you could QA everything and make your best employees 10–100x more efficient? Ben Taylor shares real use cases of business transformation and realized value in production using deep learning and discusses some of the executive conversations and behaviors Dell EMC is seeing in the market.

Carlos Escapa explains how to identify use cases for ML, shares best practices for framing problems in a way that key stakeholders and senior management can understand and support, and helps you set the stage for delivering successful ML-based solutions to your business.

There are immense opportunities to apply machine learning to investment management if you know where to look. Unlike many Silicon Valley challenges, it's not simply a matter of throwing capital and PhDs at the financial markets. Michael Weinberg explains why you must exploit domain expertise to achieve disruptive success.

Ready, set, network! Meet fellow attendees who are looking to connect at the AI Conference. We'll gather before Thursday and Friday keynotes for an informal speed networking event. Be sure to bring your business cards—and remember to have fun.

Topic Table discussions are a great way to informally network with people in similar industries or interested in the same topics.

Join fellow executives, business leaders, and strategists for a networking lunch on Thursday for AI Business Summit attendees and speakers.

Come enjoy delicious snacks and beverages with fellow AI Conference attendees, speakers, and sponsors at the Sponsor Pavilion Reception, happening immediately after the afternoon sessions on Thursday.

Don't miss AI at Night, happening on Thursday after the Sponsor Pavilion Reception.

Program chairs Ben Lorica and Roger Chen open the first day of keynotes.

In this keynote, Julie Choi explores three real-world use cases featuring a diverse set of data centric problems and Intel solutions. Julie also welcomes Ariel Pisetzky, the Vice President of IT at Taboola to discuss how AI is transforming their business as they create personalized content through their predictive recommendation engine.

Our world is one of constantly changing business environments across various business units, limited end-to-end visibility, and high alerts. Join Maitreya Natu to learn how to use machine learning to identify root causes of problems in minutes instead of hours or days by automating routine tasks without scripting or preprogramming.

Ben Lorica and Roger Chen describe the state of adoption of AI technologies and provide a glimpse into tools and trends that are poised to accelerate innovation and the introduction of new applications.

Keynote by Meredith Whittaker

Cryptocurrencies such as Bitcoin, Ethereum and others are a transformative force that could one day create an open financial network for the world. Soups manages the data science and risk team at Coinbase, one of the largest cryptocurrency exchanges in the world.

OpenAI has recently demonstrated systems capable of advanced robotic manipulation, holding their own against professionals in the massively complex game Dota 2, as well as unprecedented language understanding. Greg Brockman discusses the increasing generality of these systems and their implication for how we should think about and plan for creating safe AGI.

Much more than just a technological tool, AI has grown to the point where it often has as much influence as the people putting it to use, both within and outside the company. Kishore Durg explains why deploying AI is no longer just about training it to perform a given task. It’s about “raising” it to act as a responsible representative of the business and a contributing member of society.

With the US leading the AI revolution for decades, it would almost seem inconceivable that China could catch up. But China is rapidly catching up with the US in AI applications. Kai-Fu Lee discusses the five key factors that enabled this rapid ascension: tenacious entrepreneurs, speed and execution, tremendous capital, a big market and bigger data, and techno-utilitarian government policies.

Fireside chat with Tim O'Reilly and Kai-Fu Lee

Program chairs Ben Lorica and Roger Chen offer closing remarks.