Presented By
O’Reilly + Intel AI
Put AI to Work
April 15-18, 2019
New York, NY
Discover opportunities for applied AI
Organizations that successfully apply AI innovate and compete more effectively. How is AI transforming your business?
Be a part of the program—apply to speak by October 16.

Schedule: Models and Methods sessions

Add to your personal schedule
9:00am - 5:00pm Monday, April 15 & Tuesday, April 16
Implementing AI, Models and Methods
Location: Green Room
Francesca Lazzeri (Microsoft)
Francesca Lazzeri will walk you through the core steps for using Azure Machine Learning services to train your machine learning models both locally and on remote compute resources. Read more.
Add to your personal schedule
9:00am - 5:00pm Monday, April 15 & Tuesday, April 16
Implementing AI, Models and Methods
Location: Midtown Suite
Delip Rao (R7 Speech Science)
Delip Rao explores natural language processing with deep learning, walking you through neural network architectures and NLP tasks and teaching you how to apply these architectures for those tasks. Read more.
Add to your personal schedule
9:00am12:30pm Tuesday, April 16, 2019
Models and Methods
Location: Beekman
Gunnar Carlsson (Ayasdi)
Using Topological Data Analysis, one can describe the functioning and learning of a neural network in a compact and understandable way. This understanding results in material speedups in performance (training time + accuracy) and allows for data-type customization of neural network architectures to further boost performance and widen the applicability of the method to all data sets. Read more.
Add to your personal schedule
9:00am12:30pm Tuesday, April 16, 2019
Location: Sutton South
Garrett Hoffman (StockTwits)
Garrett Hoffman walks you through deep learning methods for natural language processing and natural language understanding tasks, using a live example in Python and TensorFlow with StockTwits data. Methods include word2vec, recurrent neural networks and variants (LSTM, GRU), and convolutional neural networks. Read more.
Add to your personal schedule
1:45pm5:15pm Tuesday, April 16, 2019
Implementing AI
Location: Sutton South
Bruno Goncalves (JPMorgan Chase & Co.)
Time series are everywhere around us. Understanding them requires taking into account the sequence of values seen in previous steps and even long-term temporal correlations. Join Bruno Gonçalves to learn how to use recurrent neural networks to model and forecast time series and discover the advantages and disadvantages of recurrent neural networks with respect to more traditional approaches. Read more.
Add to your personal schedule
11:05am11:45am Wednesday, April 17, 2019
Models and Methods
Location: Grand Ballroom West
Lise Getoor (University of California, Santa Cruz)
Much of today's data is noisy, incomplete, heterogeneous in nature, and interlinked in a myriad of complex ways. In this talk, I will describe AI methods that are able to exploit both the inherent uncertainty and the innate structure in a domain. I will describe both the benefit of utilizing structure and the inherent risk of ignoring structure. Read more.
Add to your personal schedule
11:05am11:45am Wednesday, April 17, 2019
Models and Methods
Location: Regent Parlor
Siwei Lyu (University of Albany)
In this talk, I will first briefly review the evolution of techniques behind the generation of fake media, and then introduce several projects I was involved in digital media forensics for detection of fake media, with a special focus on some of our recent works on detecting AI-generated fake videos (DeepFakes). Read more.
Add to your personal schedule
1:00pm1:40pm Wednesday, April 17, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Ameet Talwalkar (Carnegie Mellon University and Determined AI)
Hyperparameter tuning is a crucial, yet expensive, component of the ML development lifecycle. Given the growing costs of model training, we would like to leverage parallelism to tune models in roughly the same wall-clock time needed to train a single model. We propose an elegant solution to this problem, and present extensive experimental results supporting the effectiveness of our approach. Read more.
Add to your personal schedule
1:00pm1:40pm Wednesday, April 17, 2019
Machine Learning, Models and Methods
Location: Regent Parlor
Danny Lange (Unity Technologies)
Join this session to learn how to create artificially intelligent agents that act in the physical world (through sense perception and some mechanism to take physical actions, such as driving a car). Understand how observing emergent behaviors of multiple AI agents in a simulated virtual environment can lead to the most optimal designs and real-world practices. Read more.
Add to your personal schedule
1:50pm2:30pm Wednesday, April 17, 2019
Models and Methods
Location: Grand Ballroom West
vinay rao (RocketML Inc), Santi Adavani (Mr)
Growing data sizes are causing Training times to increase 5x the Moore's law. While Hardware innovations are helping, new software architectures for distributed system are needed for AI industry to solve critical problems. RocketML results show that we can build logistic regression models on KDD12 data set with ~150 Million samples on 8 Intel Xeon node cluster in < 1 minute. Read more.
Add to your personal schedule
2:40pm3:20pm Wednesday, April 17, 2019
Implementing AI
Location: Mercury Rotunda
Sumeet Vij (Booz Allen Hamilton)
The session describes the innovative application of Deep Learning to power Cognitive Conversational Agents. By leveraging Transfer Learning and deep pre-trained models for NLP, we show how Chatbots can overcome limitations of limited training datasets. In addition, we will demonstrate how Machine Learning can advances Robotic Process Automation (RPA) from “robotic” to “cognitive” automation Read more.
Add to your personal schedule
2:40pm3:20pm Wednesday, April 17, 2019
Case Studies, Machine Learning
Location: Sutton South
Vijay Agneeswaran (Publicis Sapient), Abhishek Kumar (Publicis.Sapient)
We illustrate how capsule networks can be industrialized: 1. Overview of capsule networks and how they help in handling spatial relationships between objects in an image. We also learn about how they can be applied to text analytics. 2. We show an implementation of recurrent capsule networks. 3. We also benchmark the RCN with capsule networks with dynamic routing on text analytics tasks. Read more.
Add to your personal schedule
2:40pm3:20pm Wednesday, April 17, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Sanjay Krishnan (University of Chicago)
I use my work over the last few years on building and deploying an RL-based relational query optimizer, a core component of almost every database system, as an exemplary application that highlights some of the under-appreciated challenges in Deep RL practice. Read more.
Add to your personal schedule
4:05pm4:45pm Wednesday, April 17, 2019
Case Studies, Machine Learning
Location: Sutton South
Tom Sabo (SAS), Qais Hatim (Center for Drug Evaluation and Research, U.S. Food and Drug Administration)
Drug adverse event narratives contain a wealth of information that is laborious to assess using manual methods. To improve FDA Pharmacovigilance, we apply rule-based text extraction to generate training data for deep learning models. These models improve the identification of adverse events from narrative data, enhance time-to-value, and refine sources of medical terminology. Read more.
Add to your personal schedule
4:05pm4:45pm Wednesday, April 17, 2019
Interacting with AI
Location: Regent Parlor
Kevin He (DEEPMOTION, INC.)
Digital character interaction is hard to fake–whether it’s between two characters, between users and characters, or between a character and its environment. Nevertheless, interaction is central to building immersive XR experiences, robotic simulation, and user-driven entertainment. Kevin He will discuss using physical simulation and deep learning to create interactive character technology. Read more.
Add to your personal schedule
4:55pm5:35pm Wednesday, April 17, 2019
Models and Methods
Location: Grand Ballroom West
Yishay Carmiel (IntelligentWire)
In recent years, we have seen tremendous improvements in artificial intelligence. The major breakthroughs are due to the advances of neural-based models. However, the more popular these algorithms and techniques get, the more serious the consequences of data and user privacy. These issues will drastically impact the future of AI research. Read more.
Add to your personal schedule
11:05am11:45am Thursday, April 18, 2019
Case Studies, Machine Learning
Location: Sutton South
vishal hawa (Vanguard)
While Deep Learning has shown significant promise towards model performance, it can quickly become untenable particularly when data size is short. RNNs can quickly memorize and over-fit . The presentation explains how a combination of RNNs and Bayesian Network (PGM) can improvise sequence-modeling behavior of RNNs. Read more.
Add to your personal schedule
1:00pm1:40pm Thursday, April 18, 2019
Machine Learning, Models and Methods
Location: Regent Parlor
Chang Ming-Wei (Google)
We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, BERT is designed to pre-train deep bidirectional representations by jointly conditioning on both left and right context in all layers. Read more.
Add to your personal schedule
1:50pm2:30pm Thursday, April 18, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Arun Kejariwal (Independent), Ira Cohen (Anodot)
In this talk we shall shares a novel two-step approach for building more reliable prediction models by integrating anomalies in them. Further, we shall walk the audience 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. Read more.
Add to your personal schedule
2:40pm3:20pm Thursday, April 18, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Anoop Katti (SAP)
We address understanding documents with 2D layout using machine learning. Examples of such documents are invoices, resumes, presentations etc. (in contrast to plain text documents like tweets, articles and reviews). We explore the shortcomings of the existing techniques and discuss a processing pipeline for 2D documents – the chargrid - pioneered by data scientists at SAP Read more.
Add to your personal schedule
2:40pm3:20pm Thursday, April 18, 2019
Case Studies, Machine Learning
Location: Sutton South
Alina Matyukhina (Canadian Institute for Cybersecurity)
Machine learning models are often susceptible to adversarial deception of their input at test time, which is leading to a poorer performance. In this session we will investigate the feasibility of deception in source code attribution techniques in real world environment. This session will present attack scenarios on users identity in open-source projects and discuss possible protection methods. Read more.
Add to your personal schedule
2:40pm3:20pm Thursday, April 18, 2019
Interacting with AI
Location: Regent Parlor
Behrooz Hashemian (Massachusetts General Hospital)
Artificial Intelligence has shown great potentials to revolutionize clinical medicine and health care delivery. However, incorporating these algorithms into clinical workflows faces a big challenge: convincing clinicians and regulators to trust a “black box” solution. In this talk, I present how we are making deep neural networks interpretable to provide evidences for clinical decisions. Read more.
Add to your personal schedule
4:05pm4:45pm Thursday, April 18, 2019
Interacting with AI
Location: Mercury Rotunda
Humayun Irshad (Figure Eight)
In this talk, an active learning framework with crowd sourcing approach is introduced to solve a real-world problem in transportation and autonomous driving discipline, parking sign recognition, for which a large amount of unlabeled data is available. It generates an accurate model in a cost-effective and fast way to solve the parking sign recognition problem in spite of the unevenness of the data Read more.
Add to your personal schedule
4:05pm4:45pm Thursday, April 18, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Marcel Kurovski (inovex GmbH)
Recommender Systems support decision making with personalized suggestions. They have proven useful in e-commerce, entertainment, or social networks. However, sparse data and linear models are a burden. Application of Deep Learning sets new boundaries and constitutes remarkable results. This talk shows its application on vehicle recommendations at Germany's biggest online vehicle market. Read more.
Add to your personal schedule
4:05pm4:45pm Thursday, April 18, 2019
Implementing AI
Location: Rendezvous
Jaewon Lee (LINE Corp.), Sihyeung Han (NAVER & LINE Corp)
"Until when are you going to cluster queries by yourself to manage large data corpus?" "Until when are you going to tune model hyper parameters by yourself?" I would like to introduce how to implement self-trained dialogue model by using AutoML in Chatbot within our Chatbot Builder Framework. Read more.
Add to your personal schedule
4:55pm5:35pm Thursday, April 18, 2019
Machine Learning, Models and Methods
Location: Grand Ballroom West
Maja Vukovic (IBM)
Existing AI driven automation in enterprises employ ML, NLP and chatbots. There is additional opportunity for AI Planning to drive reasoning about action trajectories to help build automation. I will demo application of AI planning for migration of legacy infrastructure to Cloud, based on real world examples and data, and discusses challenges in adopting AI planning solutions in the enterprise. Read more.
Add to your personal schedule
4:55pm5:35pm Thursday, April 18, 2019
Interacting with AI
Location: Regent Parlor
Andrew Caosun (Horace Mann School)
We propose a framework that unifies Hidden Markov Model and deep learn algorithm (RNN) with modeling components that consider long-term memory and semantics of music (LSTM and Convolution). It takes user original creation as input, modifies the raw scores, and generates musically appropriate melodies. Read more.
Add to your personal schedule
4:55pm5:35pm Thursday, April 18, 2019
Models and Methods
Location: Rendezvous
Matthew Reyes (Independent Researcher and Consultant)
This talk considers optimizing preference towards a product on a social network. The model for consumer decision-making is based on the notion of random utility. The contributions of the model are stochastic decisions that will be learned from data, and the inclusion of marketing under the control of individual companies. These contributions enable a reinforcement learning based approach. Read more.