Schedule List ViewGrid View

Vitaly Kuznetsov and Zelda Mariet compare sequence-to-sequence modeling to classical time series models and provide the first theoretical analysis of a framework that uses sequence-to-sequence models for time series forecasting.

While great strides have been made in perceptual AI (for example, in speech recognition), there's been relatively modest progress in reasoning AI—systems that can interact with us in natural ways and understand the objects in our environment. David Barber explains why general AI will be out of reach until we address how to endow machines with knowledge of our environment.

Automatic summarization is the computational process of shortening one or more text documents in order to identify their key points. Guy Feigenblat surveys recent advances in unsupervised automated summarization technologies and discusses recent research publications and datasets. Guy concludes with an overview of a novel summarization technology developed by IBM.

When building natural language processing (NLP)-based applications, you quickly learn that no single NLP algorithm can handle the wide range of tasks required to turn text into value. Amy Heineike explains how she orchestrates natural language processing, understanding, and generation algorithms to build text-based AI applications for Fortune 500 companies.

Sid Reddy shows you how to avoid the hype and decide which use cases are the best for deep reinforcement learning. You'll explore the Markov decision process with conversational AI and learn how to set up the environment, states, agent actions, transition probabilities, reward functions, and end states. You'll also discover when to use end-to-end reinforcement learning.

Cashflow is responsible for 80–90% of UK SME failure. Fluidly uses the wealth of financial data available through APIs to instantly predict cashflow. Johnnie Ball details how the company built an automated cashflow engine, explores the challenges faced in applying AI to financial data, and explains how machine learning can redefine how we think about established approaches to modeling.

Mikio Braun looks back on the past 20 years of machine learning research to explore aspects of artificial intelligence. He then turns to current examples like autonomous cars and chatbots, putting together a mental model for a reference architecture for artificial intelligence systems.

The availability of large idea repositories (e.g., patents) could significantly accelerate innovation and discovery by providing people inspiration from solutions to analogous problems. Dafna Shahaf presents an algorithm that automatically discovers analogies in unstructured data and demonstrates how these analogies significantly increased people's likelihood of generating creative ideas.

Thomas Endres and Samuel Hopstock demonstrate how to apply machine learning techniques on a program's source code, covering problems you may encounter, how to get enough relevant training data, how to encode the source code as a feature vector so that it can be processed mathematically, what machine learning algorithms to use, and more.

SURFSara and Intel collaborated as part of the Intel Parallel Computing Center initiative to advance the state of large-scale neural network training on Intel Xeon CPU-based servers. Ananth Sankar, Valeriu Codreanu, Damian Podareanu, and Steve Smith share insights on several best-known methods for neural network training and present results from tests performed on Stanford's CheXNet project.

In recent years, we’ve seen a shift from traditional vision algorithms to deep neural network algorithms. While many companies expect to move to deep learning for some or all of their algorithms, they may have a significant investment in classical vision. Paul Brasnett explains how to express and adapt a classical vision algorithm to become a trainable DNN.

Detection of moving vessels with satellite sensors is a challenging problem. Satellite imagery is expensive, covers a very small area, and can be acquired only at predefined acquisition opportunities. Natalie Fridman dives into this challenging problem and shares ISI's AI-based solution along with successful examples of detecting maritime vessels with ISI's satellites.

TensorFlow.js is the recently released JavaScript version of TensorFlow that runs in the browser and Node.js. Daniel Smilkov and Nikhil Thorat offer an overview of the TensorFlow.js ML framework and share a demo of a complete machine learning workflow, including training, client-side deployment, and transfer learning.

TensorFlow AutoGraph automatically converts plain Python code into its TensorFlow equivalent, using source code transformation. Brian Lee and Priya Gupta demonstrate how to distribute your training in TensorFlow easily across multiple accelerators and machines.

Training complex machine learning models with large amounts of data can take a very long time. Thomas Norrie explores methods for accelerating this process by distributing training across multiple accelerators and machines and leads a technical deep dive into Google Cloud’s TPU accelerators.

Pete Warden discusses the surprising effectiveness of deep learning on low-power devices.

As machine learning evolves from experimentation to serving production workloads, so does the need to effectively manage the end-to-end training and serving workflow. Kenny Song and Quentin de Laroussilhe offer an overview of TensorFlow Extended, the end-to-end machine learning platform for TensorFlow that powers products across all of Google.

Whether you’re new to machine learning (ML) or you’re already an expert, Google Cloud Platform (GCP) has a variety of tools to help you. Sara Robinson starts with the basics: how to use a pretrained ML model with a single API call. She then demonstrates how to customize a pretrained model with AutoML. Sara concludes by explaining how to train and serve a custom TensorFlow model on GCP.

Susan Etlinger explores how AI fundamentally changes the relationship between people and businesses, lays out its risks and opportunities, and demonstrates emerging best practices for designing customer-centric and ethical products and services.

Max Gadney and Sabih Ali explore some of the ways designers and product teams are designing systems with transparency, trust, and privacy in mind.

AI is a key innovation accelerator for digital business transformation. To help you with your strategic roadmap, Philip Carnelley shares IDC's research into the AI market across hundreds of European organizations and explains why organizations should establish a digital platform based on big data, AI, and cloud technologies, with an intelligent core, as part of their transformation strategy.

Advertising in China is on the frontline of AI adoption and innovation. Join Bessie Lee and Ching Law for a conversation on how AI is changing advertising. You'll hear how China's white-hot AI advertising applications can serve as roadmaps and spark ideas in other industries and how companies like Tencent are improving performance by leveraging AI technology.

What does a workforce augmented by digital humans look like? Marie Johnson shares the story of the creation of Nadia, the world’s first digital human for service delivery. Drawing on her experience developing the concept and leading the delivery, Marie presents a framework to help leaders meet exponential changes across industries augmented by digital humans, including healthcare.

As organizations turn to data-driven strategies, there's been increasing interest in creating AI centers of excellence (COEs). Benjamin Wright-Jones and Simon Lidberg take you through the building blocks of a center of excellence and describe the value for organizations embarking on data-driven strategies.

Understanding user listening behavior is essential for personalizing music listening experiences on Spotify. Mounia Lalmas explains how Spotify uses machine learning recommenders that take into account what and how users consume playlists and the rich diversity of playlist experiences.

Recent research has shown that training for quantization can lead to large gains in energy efficiency, and embedded runtime packages like TensorFlow Lite and Caffe2Go offer portability over a number of platforms. Cormac Brick asks, Why can't we have both performance and portability? Cormac explores industry challenges and details the progress needed to close the portability-performance gap.

Even in the age of big data, labeled data is a scarce resource in many machine learning use cases. Florian Wilhelm evaluates generative adversarial networks (GANs) when used to extract information from vehicle registrations under a varying amount of labeled data, compares the performance with supervised learning techniques, and demonstrates a significant improvement when using unlabeled data.

On his journey to the top spot at Kaggle, Marios Michailidis noticed that many of the things he does to perform competitively in data challenges could be automated. Marios shares lessons learned from his Kaggle experience and shows how you can achieve competitive performance in predictive modeling tasks automatically, using H2O.ai’s Driverless AI—an AI that creates AI.

Dramatic progress has been made in computer vision. Deep neural networks (DNNs) trained on millions of images can recognize thousands of different objects, and they can be customized to new use cases. Vanja Paunic and Patrick Buehler outline simple methods and tools that enable users to easily and quickly adapt Microsoft's state-of-the-art DNNs for use in their own computer vision solutions.

Christopher Cho details how to leverage Kubernetes and the mighty Kubernetes APIs to build a complete deep learning pipeline, from data ingestion and aggregation to preprocessing and ML training to serving. Along the way, Christopher covers Kubeflow, a Google open source solution for managing machine learning with TensorFlow in a portable, scalable manner.

How can AI impact national security? Collaborating with the UK Home Office Counterterrorism Unit, ASI Data Science built a tool that removes extremist propaganda from the web. Drawing on this experience, Marc Warner discusses the role of AI in the fight against terror and explains how shared access to this technology may be part of the answer.

The common perception of applying deep learning is that you take an open source or research model, train it on raw data, and deploy the result as a fully self-contained artifact. The reality is far more complex. Nick Pentreath shares lessons learned building a deep learning model exchange and discusses the future of standardized cross-framework deep learning model training and deployment.

Chris Boyd and John Wiley explain how the Wall Street Journal uses machine learning and a proprietary algorithm to predict the likelihood for someone subscribing, which in turn dictates the paywall experience that customer receives.

We're in the year of the AI fake out. "Fake news" is the order of the day, as nebulous chatbots have become significant political actors. Startups peddle robotically handwritten notes and algorithmically personalized gifts for our loved ones. Soon we won't even be able to tell if a customer service agent is a real person. Aileen Nielsen asks, How should we redefine intelligence as fakes flourish?

With the more applications of machine learning-based applications, the complex algorithms that automate behaviors can get out of control. Ira Cohen explains how to catch problems and glitches early on by using machine learning algorithms to monitor these algorithms for anomalous behavior.

Google's AIY Projects kits bring Google's machine learning algorithms to developers with limited experience in the field, allowing them to prototype machine learning applications and smart hardware more easily. Alasdair Allan walks you through setting up and building the kits and demonstrates how to use the kits' Python SDK for machine learning both in the cloud and locally on a Raspberry Pi.

Andrew Trask details the most important new techniques in secure, privacy-preserving, and multiowner governed artificial intelligence and offers a demonstration of the OpenMined project.

Christopher Nguyen shares lessons learned implementing multiple AI commercial projects at Panasonic. Along the way, Christopher discusses a number of use cases at various stages of implementation maturity and explains what AI really means today in enterprise products, where the key opportunities are, their impact, and key success factors in the adoption of AI across the enterprise.

Transfer learning allows data scientists to leverage insights from large labeled datasets. The general idea of transfer learning is to use knowledge learned from tasks for which a lot of labeled data is available in settings where only little labelled data is available. Lars Hulstaert explains what transfer learning is and demonstrates how it can boost your NLP or CV pipelines.

Blockchain technologies offer new internet primitives for creating open and online data marketplaces. Roger Chen explores how data markets can be constructed and how they offer a shared resource on the internet for AI-based research, discovery, and development.

Machine learning and AI models now outperform humans on many tasks. However, sending sensor data up to the cloud and back is too slow for many apps and autonomous machines. Jameson Toole explains why developers seeking to provide seamless user experiences must now move their models down to devices on the edge, where they can run faster, at lower cost, and with greater privacy.

Zhipeng Huang explains how resource representation (RR) works with various intermediate representation (IR) technologies to help achieve the democratization of AI.

Multitask learning is an approach to problem solving that allows supervised algorithms to master more than one objective in parallel. Ryan Micallef shares a multitask neural net in PyTorch trained to classify news from several publications, which highlights distinct language use per publication enabled by the analysis of task-specific and agnostic representations part of multitask networks.

Deep learning for time series prediction has made rapid progress in the past few years, but performance still greatly lags that of other intelligence tasks. Aileen Nielsen offers an overview of the state of the art in 2018, covering the hottest new architectures, emerging best practices for RNN training, and long overdue standard metrics to measure and compete on neural network prediction.

When you train a model on private data, how much of that information does the model retain? Katharine Jarmul reviews research on attacks against models to extract training data and expose potentially sensitive information. Katharine then shares potential defenses as well as best practices when training models using private or sensitive data.

Peter Cahill explains why Wavenet will be the next generation of recognition, synthesis, and voice-activity detection.

Archisman Majumdar and Jai Ganesh describe the effects of AI techniques on frontend GUI development—specifically, the use of automatically generated code and architecture from text descriptions—and share deep learning techniques for text-to-image creation and template-to-code generation, along with cloud technologies in automated deployment, management, and scaling of such applications.

Program cochairs Ben Lorica and Roger Chen open the second day of keynotes.

Kristian Hammond walks you through an approach to bring AI into the enterprise, based on the functional, business aspects of AI technologies. Kristian maps out simple rules, useful metrics, and where AI should live in the org chart, laying out the route you should follow to make good on the promise of the technologies of intelligence.

Fireside chat with Marc Warner and Louis Barson

Jason Knight offers an overview of the state of the field for scaling training and inference across distributed systems from a practitioner's point of view. Along the way, Jason dives deep into available tools, resources, and venues for getting started without having to go it alone.

Supasorn Suwajanakorn discusses the possibilities and the dark side of building artificial people.

Why do businesses fail at machine learning despite its tremendous potential and the excitement it generates? Is the answer always in data, algorithms, and infrastructure, or is there a subtler problem? Will things improve in the near future? Cassie Kozyrkov shares lessons learned at Google and explains what they mean for applied data science.

Drawing on the McKinsey Global Institute's groundbreaking research, Michael Chui explores commonly asked questions relating to AI and its impact on work. Michael also previews new research showing that despite the rapid pace of AI adoption, much foundational work in enterprises remains to be done to capture value at scale.

Program cochairs Ben Lorica and Roger Chen close the second day of keynotes.

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.

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