DATAx

10:45am – 11:15am

PANEL DISCUSSION: Preserving the Virtue of Your Data – (Track 1)

Do not assume that everyone in your company values data in the same way. Each department has its own unique relationship with data. Implementing clear processes around data is the only way to safeguard your company’s most valuable asset. In this session, topics of discussion will include ▪ Incorporating data governance into your business strategy ▪ Developing a culture of data citizens ▪ Processes for preserving data quality

10:45am – 11:15am

USE CASE: Demystifying Applied ML – Building Frameworks & Teams to Operationalize ML at scale – (Track 2)

Comprehending how to match the deep knowledge of a subject matter expert to the technical application of ML programs remains a major barrier to applied ML in the workplace. Join us as we examine best practices on how to scale operationalizing ML to solve complex problems. In this session, topics of discussion will include: ▪ Outlining your business Objectives: defining goals and success criteria. ▪ Providing the right technology and tools ▪ Structuring and training your teams

Dr. Nels Lindahl Director, Clinical Decision Systems
CVS Health

10:45am – 11:15am

USE CASE: How Strava is Using Data Visualization to Fuel Growth – (Track 3)

The Strava dataset grows by millions of activities a day and comprises trillions of GPS data points; this quantity of data is gold for businesses. Discover how user data at Strava is transformed into innovative visualizations of human performance: heatmaps, fitness stats, mobile routes, and even urban planning/mobility. Hear first-hand how they are able to scale visualizations, and detail how data sets are analyzed and transformed into business insights the company uses to fuel its growth. In this session, topics of discussion will include: ▪ How to effectively scale visualization ▪ Using Data Visualization for increase growth ▪ Extracting quality data from large datasets

Cathy Tanimura Senior Director, Analytics & Data Science
Strava

10:45am – 12:25pm

WORKSHOP: Exploring the Race to Capitalize on Untapped Big Data Reserves to Develop Smarter, Data-driven Decisions – (Track 4)

AI solution reliance on modern socio-economic trends is one of the most provocative factors for data-driven analytics. Businesses are relying on insights extracted from Big Data, substantiating data as the new form of currency. This paradigm shift has corporations racing to capitalize on their untapped big data reserves to develop smarter, data-driven decisions capable of improving every business aspect. The complexity of moving to data-driven frameworks is often underestimated. HPE’s AI-driven Full Stack Data Science program works strategically to develop custom, high performing, end-to-end, real-time analytics solutions focused on facilitating client’s success making intelligent, data-driven decisions from their amassed data In this session, topics of discussion will include: • Learn how AI-Driven Full Stack Data Science pipelines power the Extreme-scale Market • Explore the real costs associated with Free and Open Source Software • Understand how Commercial Products offset many development complexities Hosted by Micro Focus

Theresa Melvin Chief Architect, AI-Driven Big Data Solutions
Hewlett Packard Enterprise

11:20am – 11:50am

Self Service Enablement – Empowering Data Discoveries Across the Organization – (Track 1)

Achieving the level of a data driven organization opens up a new world of possibilities for your organization. With a stable data governance strategy in place and a workforce fluent in data you are able to empower your workforce to drive data focused initiatives across the organization. In this session, topics of discussion will include: ▪ Defining your data governance strategy ▪ Establishing a level of data fluency within your organization ▪ Training on the tools and techniques required to effectively analyze data

11:20am – 11:50am

How to Effectively Accelerate ROI on Your AI Investments – (Track 2)

Every organization wants to use data as its core advantage, but struggle with scaling and accelerating their data initiatives. Many of the challenges can be attributed to the fact that the data, and its applications are advancing faster than the technologies used to process and design them. Join us as we dive into the root cause of many common data challenges and incorporate applied patterns to solve them. In this session, topics of discussion will include: ▪ How to design efficient data models that will enhance the performance of your AI/ML tools. ▪ How to successfully use applied patterns to address the root cause of your data challenges ▪ Real-World solutions to scale and accelerate AI/ML processes

Ronak Shah Head of Data Engineering
Coursera

11:20am – 11:50am

PANEL DISCUSSION: Dirty Data – Strategies for Improving the Quality of Your Data – (Track 3)

Are you confident in your organizations data? If the answer is NO, then you have a major problem. High quality data can increase business performance for your organization and validate key business decisions Learn best practices for protecting the quality of your data. In this session, topics of discussion will include: ▪ Increasing business performance ▪ Improving data quality ▪ Protecting our data

11:55am – 12:25pm

USE CASE: Business Strategies to Burst the Bias Bubble – (Track 1)

Collaborative decision making under uncertainty happens daily in business. Machine learning obscures bias behind the complexity of its algorithms. Bias is corrosive because it hides in metrics and the systems we use to make decisions every day. It gives us false confidence that our decisions are data driven when in reality, they are bias driven. Even data scientists often don’t understand the data and models well enough to prevent biased outcomes. Autonomous cars are 5% more likely to hit PoC because their training data wasn’t diverse enough. Amazon abandoned their automated candidate screening system because it was biased against protected classes. Join us as we dive into what you can do to control bias within your organization. In this session, topics of discussion will include: ▪ Creating a culture that strives to reduce bias ▪ Using Machine Learning systems to reduce uncertainty ▪ Reducing uncertainty around key decisions by controlling bias

Vin Vashishta Data Scientist, Strategist, Author
V2 Machine Learning Consulting

11:55am – 12:25pm

USE CASE: Explainable Artificial Intelligence (XAI) – (Track 2)

Deep learning algorithms have achieved high performance accuracy in many complex domains. Due to the nested non-linear structure of deep learning algorithms, these highly successful models are usually applied in a black-box manner, i.e., no information is provided about what exactly causes them to arrive at their predictions Explainable Artificial Intelligence (XAI) creates interpretable models while maintaining a high level of learning performance, thereby enabling users to understand, appropriately trust the underlying models. The session will provide a comprehensive overview of XAI concepts along with three use case demonstrations across biomedical, natural language processing and security applications. In this session, topics of discussion will include: • Creating interpretable models while maintaining high accuracy • How to leverage explainability to increase trust in model decisions? • How XAI enhances understanding of “black box” models & feature engineering?

Sherin Mathews Senior Data Scientist
Mcafee

11:55am – 12:25pm

Augment, Don’t Automate: Drawing Insights from Customer Feedback Using Natural Language Processing – (Track 3)

Companies are frequently faced with large amounts of unstructured text data, like forum comments or product reviews. Important trends can emerge in these datasets, but it can be time-consuming to read through comments, and keyword matching frequently misses critical nuances. We’ll discuss how we’ve approached this problem at Google using Natural Language Processing, with examples of the approach applied to open datasets. We’ll explore how this fits into the ML project lifecycle, with examples of common pitfalls. Finally, we’ll highlight how to use this technology as part of a “human in the loop” approach to supercharge your existing team members.

Peter Grabowski Software Engineering Manager
Google

12:30pm – 1pm

AI + Humans Working Together to Drive Revenue – (Track 1)

When it comes to AI collaboration is key. The fear that humans will be replaced by the machines is still very prevalent in the workforce. The reality is AI is a tool and humans will always be needed to manage the tools. Data leaders must spearhead the change in mindset around the implementation of AI. Working together machines will provide the consistency and speed to uncover patterns in real-time and humans will provide the logic to extract insights that will help companies make better informed decisions. In this session, topics of discussion will include: ▪ Fear of the machines ▪ Scaling AI across an organization ▪ Upskilling your workforce

12:30pm – 1pm

USE CASE: Using Machine Learning to Get Personal – (Track 2)

Learn how GSN Games uses machine learning to integrate behavioral data, survey data and psychographic data to generate game-play personas for their players. These personas are used to help inform and validate product features and in-game experiences, to make their games more fun and rewarding for players In this session, topics of discussion will include: • Data goes beyond behavioral actions • Customers have different aspirations and motivations when they engage with products • A successful product is keenly informed by customer aspirations

Andy Veluswami Vice-President of Data Engineering & Data Science
GSN Games (a division of Sony Pictures Entertainment)

12:30pm – 1pm

Effective Algorithms to Gain a Competitive Advantage – (Track 3)

The right algorithms allow companies to disrupt industries and change the way consumers spend their time and money. Just hiring a data science team is not enough to develop effective algorithms. Sometimes, data science teams struggle to find the right data for the problem, struggle to solve the problem well enough, or struggle to implement the solution. In this session, topics of discussion will include: • Finding the right problems & Setting the right, realistic objectives • Testing and validating results • Engaging stakeholders to use the algorithms & customers to understand and trust the algorithms

Elizabeth (Betsy) Barton Director of Data Science
InHome Delivery, Walmart

2pm – 2:30pm

USE CASE: Getting Your Data House in Order: Critical Issues in Advancing Data Governance – (Track 1)

Data is the asset of the present – and the future. Whether it’s generating insights, driving innovative product development or improving decision making, the need for and dependence on big data has never been greater. Data is a critical strategic asset for any company looking to leverage the power of AI, machine learning, and other advanced analytics. However, for companies that were not “data-first,” implementing a data governance program that both maximizes the quality of their data and addresses the obligations created by the changing regulatory landscape remains a challenge. This session will explore the critical legal and practical issues a business must consider as it develops and implement compliant and effective data programs. In this session, topics of discussion will include: ▪ How to structure a data governance program that addresses their current regulatory obligations to comply with new privacy and security obligations ▪ Best practices in creating a data-focused business culture ▪ Real-world examples of trends, challenges, opportunities and developments to come in data collection and use

Jessica B. Lee Partner, Cp-Chair, Privacy, Security & Data Innovations
Loeb & Loeb

2pm – 2:30pm

PANEL DISCUSSION: Is AI the Right Fit for Your Business? – (Track 2)

There is still a lack of understanding of how AI & ML should be used within organizations. Many companies have spent considerable amounts of money (blown budgets) on AI/ML technology and have not seen any ROI on these investments. Leaders are tasked with sourcing new technology, but they need to know that the technology is the best fit for the business. This session will help you manage your expectations around AI/ML and provide you with realistic goals for how AI/ML can fit your business objectives. In this session, topics of discussion will include: ▪ AI on a shoestring budget ▪ Sizing AI to fit your business ▪ Managing stakeholder expectations around AI

2pm – 2:30pm

USE CASE: How to Save your Company from Drowning in Unimpactful Insights – (Track 3)

With many organizations now relying on a multitude of data sources to inform their decision-making, the expectation has now been set on stakeholders and leaders to depend on, and act on, a perpetually growing technology stack and volume of data. Despite the competitive advantage that clean data can bring to an organization, there also exists a real risk of drowning in an abundance of interesting, but ultimately unimpactful, data and insights. In this talk, we will go through two detailed use-cases of how Getty Images combined Data Science and business needs to build a comprehensive suite of data-based frameworks and products to track, monitor and alert on e-commerce performance, and to help understand how customers consume our vast collection of imagery content. In this session, topics of discussion will include: • The importance of building business-agnostic frameworks and applications • How to understand customer behavior using NLP and unsupervised learning • How to leverage time series modeling to move from a reactive to proactive state when monitoring e-commerce performance

Thomas Vincent Head of Data Science
Getty Images

2pm – 3:30pm

Data Visualization – (Track 4)

In this immersive workshop you will get hands on training on some of the latest tools in data visualization. Learn how to transform data into knowledge that is relevant for decision making. Topics of discussion will include: ▪ Visual Data Storytelling ▪ Which tools are most effective for your industry? ▪ Extracting the most relevant information to showcase

2:30pm – 3pm

USE CASE: A Guide to Building a Data Driven Culture – (Track 1)

When it comes down to it, organizations struggle with becoming data driven and making it a part of their culture. Hiring a team of data scientists does not fix this problem outright. Success is dependent upon building the right processes, democratizing data and resetting the mentality across the organization. These problems are often more difficult to solve than the tech and math related to data science. Join us in the discussion of a how to guide for building a data driven culture. In this session, topics of discussion will include: ▪ Understand the data maturity model and where to start. ▪ Get an overview of processes and people requirements to drive a data driven culture. ▪ Understand the citizen data scientist movement and the realistic expectations of people outside of data science

Daniel Gremmell Vice President, Data Science
Plated

2:30pm – 3pm

PANEL DISCUSSION: Deploying Machine Learning in the Wild – (Track 2)

Join in on this discussion as representatives from some of today’s leading-edge company’s share how they are developing and deploying ML applications with success. Hear the backstories behind the journey; pitfalls, abandoned projects, small wins and successful processes that worked. In this session, topics of discussion will include: ▪ Managing stakeholder’s expectations when developing machine learning ▪ Structuring your team for effective ML implementation ▪ Preparing your data for algorithm training – Do you know what problems you want to solve?

Elizabeth (Betsy) Barton Director of Data Science
InHome Delivery Walmart

2:30pm – 3pm

USE CASE: The Intuition Behind Machine Learning in Marketing – (Track 3)

Since the year 2013 important breakthroughs and advances in technology have made it possible to run sophisticated predictive models capable of classifying images, text, and sound. Technology has brought to reality self-driving cars, chat bots and a host of other AI powered devices. In this session we will present key insights that will help you make AI/ML more beneficial to your marketing efforts. Through real world case studies, the session will demystify the core technologies around ML and illustrate how to successfully apply the technology. In this session, topics of discussion will include: ▪ How to think and interpret predictive models ▪ What metrics we use to evaluate models ▪ Specific case studies in optimization, channel attribution

Mario A. Vinasco Director BI and Analytics
Credit Sesame

3pm – 3:30pm

PANEL DISCUSSION: Systems Integration – Getting Your Systems to Work in Harmony – (Track 1)

As companies grow so does the technology used to support them; with that growth comes a myriad of systems implementation. Many companies are currently struggling with getting their systems to communication. This lack of communication has made it challenging to easily access all aspects of captured data. Discover real-world examples of how some companies are navigating this communication roadblock. In this session, topics of discussion will include: ▪ Navigating system integration ▪ Growth and technology ▪ Digital transformation

3pm – 3:30pm

USE CASE: Using ML and AI to Produce Real-Time, Scalable, Informed Decisions – (Track 2)

The ability of ML and AI to provide real-time, scalable decision-making support to executive strategy is changing the business landscape. In the healthcare sector, patient privacy, data interoperability, and high stakes patient health ramifications intertwine to create a challenging environment in which to use advanced computation techniques. Join us to learn how Johns Hopkins Healthcare’s model, Callisto has been instrumental in predicting suitable health management programs for THEIR patients that deliver concrete return on the overall health of the patient. In this session, topics of discussion will include: ▪ The unique challenges within the healthcare sector that other industries can benefit by understanding ▪ How the runaway cost structures can be fundamentally inverted by targeting pain points with focused ML/AI ▪ How advancements in AI has allowed the healthcare sector to leapfrog into the 21st century as a fully-integrated player in the cyber community

Romy Hussain Director of Healthcare Economics and Data Science
Johns Hopkins Healthcare

3pm – 3:30pm

USE CASE: Streaming (Real Time) Analytics – (Track 3)

Join us as we unbox how TD Banks is taking advantage of streaming analytics in real time. Through real world case studies, you’ll learn how they are using this quick time sensitive process as a competitive edge. In this session, topics of discussion will include: ▪ Five Levels of Streaming Analytics Maturity ▪ Advantages and Disadvantages of streaming analytics in real time ▪ Supportive open source technology

Bharti Bhardwaj Delivery Director
TD Bank

10:45am – 11:15am

EXECUTIVE ROUNDTABLE DISCUSSIONS: Leadership Strategies for Today and Beyond – (Track 1)

These roundtable sessions provide a great opportunity for data leaders to exchange ideas around common industry challenges. These intimate facilitated off the record table discussions give leaders the opportunity to problem solve with peers and discuss strategies around implementation and the future of data and AI. In this session, topics of discussion will include: ▪ Exchange of ideas around how you and your fellow industry peers are implementing data strategy within your organizations ▪ Exchange of ideas around current data challenges ▪ Exchange of ideas around leadership and team design

10:45am – 11:15am

PANEL DISCUSSION: Ethical Risks in AI – (Track 2)

There are many factors that can bring into question the ethics of your data programs, such as bias, inconsistency, security, and transparency to name a few. The existence of these risks affects whether your AI systems should be trusted. Join us for a conversation around ethics and the role it plays in your AI strategy. In this session, topics of discussion will include: ▪ Defining an AI strategy ▪ How to effectively build trust in your data ▪ Bias in algorithms

Bhargav Raman, CEO
Medpixels

10:45am – 11:15am

USE CASE: Augmenting Customer Experience with ML in eCommerce – (Track 3)

The eCommerce industry has been completely transformed by the advancements in AI and ML. The ability to recommend and personalize products have allowed the industry to grow rapidly. With this growth the ability to effectively harness data from digital platforms have become increasingly challenging. In this session, topics of discussion will include: ▪ The advantages and disadvantages of personalized vs. non-personalized recommendations ▪ Challenges involved with harnessing data for customer recommendations on digital platforms ▪ Primary considerations when deciding to implement a product recommendation engine

Asma Farooq Machine Learning Product Lead
Ebay

10:45am – 12:45pm

A Practical Introduction to Natural Language Processing – (Track 4)

A practical introduction to (NLP) Natural Language Processing is becoming an essential skill for organizations to gain a competitive edge. By enabling services like chatbots, text-based decision support systems, semantic analysis, threat detection and question answering systems NLP tools are being used to help people get insights from unstructured data and operate more effectively. For this reason, it has become an essential tool for many new business functions. In this immersive workshop you’ll be introduced to NLP techniques and tools with their real-world applications. In this session, topics of discussion will include: ▪ Learn and understand popular NLP approaches with interactive examples ▪ Understand how to clean and handle unstructured text data ▪ Complete NLP tasks including text classification, sentiment analysis, text clustering, topic modeling and semantic visualizations

Carlo Lipizzi PhD, Graduate Engineering Management & Systems Analytics Program Director School of Systems and Enterprises
Stevens Institute of Technology

11:20am – 11:50am

PANEL DISCUSSION: Structuring Data Teams to Maximize Impact – (Track 1)

A strong data team affords you the ability to effectively solve for business problems and leverage the rich resources within data to influence your company’s direction for the future. Lean in on this discuss with Data Science leaders as they share their struggles and successes in organizing data science teams. In this session, topics of discussion will include: ▪ How to best organize data teams within an organization ▪ The key to hiring and retaining top notch talent ▪ Team fluency around financial and business metrics

Meghan Anzelc Head of Data & Analytics
Spencer Stuart

11:20am – 11:50am

ASK THE EXPERTS ROUNDTABLE DISCUSSIONS: Data & AI Wellness Check – (Track 2)

These intimate solutions focused roundtables offer you a great opportunity to check in with industry leaders about the state of y our data and AI health. These off the record discussions are great ways to get guidance for your most pressing questions around your data and AI journey. In this session, topics of discussion will include:  Current data and AI challenges you and your in dustry peers are fa cing  Sourcing AI for your business needs ▪ How to incorporate AI/ML in to the your organization

11:20am – 11:50am

USE CASE: How Tencent Games is Successfully Shaping Business Strategies & Maximizing Income – (Track 3)

At Tencent Games, all decisions are data driven. Data mining techniques are applied to huge data sets collected from 200+ games to provide end to end big data analytic solutions. In this session , you will learn how they apply various predictive modeling techniques throughout the customer life cycle to improve user experience and optimize marketing. They will present use cases on new user acquisition, user monetization, user retenti on , churn pre dictions, and item recommendation In this session, topics of discussion will include:  How big data transforms the gaming industry  Data science solutions at scale throughout user life cycle  How to choose an appropriate algorithm for a sp ecific problem

Qiaolin Chen Director of Data Science
Tencent Games

11:55am – 12:25pm

USE CASE: Making Data Science – Scoping, Sizing, Hiring & Retaining an Internal Data Science Capability – (Track 1)

Developing an internal data science capability requires a major cultural shift and a strategic mapping process which align your business objectives. In order to be effective, you need technical infrastructure that support new processes and organizational structure that can alter business practices that create measurable efficacy. In this session, topics of discussion will include: ▪ Opportunities for data science to identify and prioritize value for your organization ▪ Recruiting, retaining, and goal setting for team members ▪ Budgeting and hiring for business objectives

Haftan Eckholdt Chief Data Officer & Chief Science Officer
Understood.org

11:55am – 12:25pm

USE CASE: Measuring Incremental Value of a Business Using Machine Learning Algorithms – (Track 2)

Wayfair launched the CastleGate program, a leading drop-ship operation in 2014, helping 300+ suppliers boost sales by delivering products to customers faster, safer, and at a lower cost than ever before. As they expanded their supplier partnership footprint in CastleGate, it was critical for them to quantify incremental dollar impact of entering a product in the program. This translated into a challenge for their analytics team due to the absence of historical experimental design. Join us for this session as they share the journey of this project. In this session, topics of discussion will include: ▪ The importance of showing incremental dollar impact ▪ How to push forward in the absence of historical information ▪ How to realistically boost sales

Ankit Mangal Associate Director, Web Analytics
Wayfair

11:55am – 12:25pm

PANEL DISCUSSION: How to Creatively Convey Your Data Discoveries to Non-Technical Stakeholders – (Track 3)

Gaining support from internal stakeholders is the life or death of many initiatives. Often, projects are rejected because stakeholders don’t have a clear understanding of its possibilities. Discover effective ways to use data visualization to influence stakeholder buy-in. In this session, topics of discussion will include: ▪ Using data visualization as a selling tool ▪ How to win over non-technical stakeholders ▪ Data Visualization tools

12:30pm – 1pm

Leading with Data – Creating a Business Strategy Where Data is King – (Track 1)

In order to identify how data can be instrumental in solving key business priorities you much first clarify the business drivers where data can deliver value. Your strategy should clearly demonstrate data’s ROI. As data continues to be imbedded in all industries the only way to maintain relevance within your industry is to create a business strategy lead by data. In this session, topics of discussion will include: ▪ Where do your data decisions originate top down or bottom up? ▪ Where are you in your data journey? ▪ Clarifying business objectives

12:30pm – 1pm

USE CASE: Managing Bias in Predictive Models – (Track 2)

AI-based approaches to prediction are often seen as an unbiased decision-making process. However, algorithms have frequently been found to have biased predictions. Whether due to statistical issues in data collection or labeling, or because of human bias in training data, these issues can impact individuals at a tremendous scale. In this session, we will highlight examples of biased predictive algorithms and cover primary sources of bias in the development of predictive models. We will also discuss best practices that can be used to assess for, and hopefully limit, bias in predictive models. In this session, topics of discussion will include: ▪ Understand the limitations of predictive models and the risk for bias in AI ▪ Describe key risks in data or approach to model development that can lead to bias in predictive algorithms ▪ Develop internal methods to assess for and limit

Wade Schulz, MD, PhD Assistant Professor of Laboratory Medicine & Computational Healthcare Researcher
Yale University School of Medicine

12:30pm – 1pm

PANEL DISCUSSION: How to Achieve a Seamless Digital First Experience – (Track 3)

In today’s digital marketplace the majority of the digital audience are anonymous. This makes it extremely difficult for many customer facing organizations to get a good understanding of its customer base. Find out how some of the biggest brands are curating effortless digital experiences that is allowing them to attract and retain customers at high rates. In this session, topics of discussion will include: ▪ Collecting data around the digital experience ▪ Best practicing in digital analytics ▪ Creating a seamless digital experience

2pm – 2:30pm

USE CASE: “Data for Good” an Essential Building block for a Healthy Data Culture – (Track 1)

The “Data for Good” movement is increasingly becoming a critical component to many organization’s corporate social responsibility initiative. The benefits of this movement include improved employee engagement, increased brand visibility and greater community engagement. Viz for Social Good is one of the many organizations whose mission is to help non-profit organizations understand their data through attention-grabbing and informative visuals. Discover how to harness the talents of data and analytics specialists to impact social change. In this session, topics of discussion will include: ▪ Benefits of incorporating Data for Good into your organization to improve data culture ▪ How to use your expertise and talents to impact humanity ▪ Defining Data for Good and methods of engagement within your organization

Vanitha Lucas Board Member
Viz for Social Good

2pm – 2:30pm

PANEL DISCUSSION: Human Centered AI – (Track 2)

Join us for an in-depth look at how machines are taking lessons from their human counterparts. Discover how machines are becoming more humanized in order to be effective at finding the best insights. In this session, topics of discussion will include: ▪ Human-centered design ▪ Human behavior patterns ▪ Effective implementation of emerging technology

2pm – 2:30pm

USE CASE: Application of Image Based KNN and Anomaly Detection to Product Type Classification – (Track 3)

Application of Image Based KNN and Anomaly Detection to Product Type Classification In e-commerce product catalog, product types have a significant influence on customer journey starting from product discovery. Image-based classification approaches do not have high enough precision by itself to be used in production. On the other hand, text-based classification models are challenged by noisy input data, flattened product type hierarchy, and certain kitchen sink product types. By combining image based KNN with anomaly detection, the speaker will demonstrate how to detect product type errors in an unsupervised manner. In this session, topics of discussion will include: ▪ How to use anomaly detection to validate golden data set ▪ How to optimize recall/precision of image based KNN ▪ How to build end-to-end misclassification detection pipeline for large scale catalog

Binwei Yang, PhD Distinguished Engineer, Merchant Technology Data Science,
Walmart Labs

2pm – 3pm

WORKSHOP D: The Neural Network Playground – (Track 4)

In this hands-on workshop attendees will have an opportunity to build their very own neural network. We’ve created a safe space for you to play around and test your skills and theories. Whether it’s your first time or you are a pro; I’m sure you’ll gain great insights from your conference colleagues. In this session, topics of discussion will include: ▪ Popular tools for building networks ▪ Best practices in data science and infrastructure ▪ Effective uses of neural networks

2:30pm – 3pm

PANEL DISCUSSION: Creating a Data Science Center of Excellence – (Track 1)

There are so many aspects that go into creating a data science CoE. Funding, who will spearhead the effort and building the best teams. In this session you’ll hear how companies are stepping up to the challenge of taking charge of their data. In this session, topics of discussion will include: ▪ Funding your Data Science CoE ▪ Data Culture ▪ Curating a team of unicorns

2:30pm – 3pm

USE CASE: Developing Data Science to Add Value – (Track 3)

The goal of every data science team is to make a significant impact on the business. Teams work hard building some amazing and extremely useful tools that never see the light of day and their potential impact never realized. Join us as we uncover how to organize the development process to keep engagement high and demonstrate added value along the way. In this session, topics of discussion will include: ▪ Obstacles faced by data scientists their data projects ▪ How to restructure development, get accurate feedback and demonstrate value ▪ How to keep stakeholders involved during project journey. From exploration to implementation

Morgan Cundiff Data Scientist
ShopRunner