Advanced Analytics & Data Science (Day #2)

Join experts from Lionhead Studios, King.com, Nordeus, Jagex, Mind Candy and Scientific Revenue for a day-long course on the state-of-the-art of analytics techniques for understanding and enhancing player experience and design choices — whether for games or apps — using combinations of machine learning and clever modeling tricks!

Now in its 7th edition, nucl.ai brings together hundreds of developers from creative industries and interactive media to resolve the biggest technical and design problems with Artificial Intelligence. One of the event's tracks focuses specifically on “Advanced Analytics & Data Science” for an entire day. (Day #2: Tuesday, July 21st)

17 Day Tickets Left! Attending Companies & Studios

  • King.com
  • Square Enix
  • Lionhead Studios
  • Riot Games
  • Creative Assembly
  • Ubisoft Studios
  • Wooga
  • Unity Technologies
  • CD Projekt Red
  • Nordeus
  • Scientific Revenue
  • Jagex
  • Microsoft Game Studios
  • Natural Motion
  • Mind Candy
  • Epic Games
  • Zynga
  • CCP Games
  • ...

I had a great time attending the AI Conference and met so many amazing and nice People there. Once again thank you for giving us students the Oportunity to visit this great Conference!

Jonas Jentsch

Gamedesign Student

Mediadesign Hochschule, GERMANY

The most informative and conference I've been too. Great value and great people. I'll be back.

Chris J. Rock

Wooga, GERMANY

This year's Game/AI Conference was the best one I've gone to so far. The programme was a great balance of high- and low-level, tech and design, as well as academia and industry. The organization was flawless and the location was great.

Jurie Horneman

Independent game designer / programmer

Awesome conferences, where professionals are sharing their knowledge, and their passion. It's quickly going to be the reference in the AI community.

Xavier Dolci

Lead AI Programmer

Ubisoft, FRANCE

This conference was a real deal — cutting edge presentations and experienced speakers. My take away list from this one is impressive! I feel this event can really push Game AI forward!.

Mieszko Zielinski

AI Programmer

People Can Fly, POLAND

I thought the conference was a great success. I learned a lot from the speakers at the conference, as well as interacting with the attendees.

Frederik De Caster

AI Programmer

Creative Assembly, ENGLAND

Thank you for this amazing conference! Great speakers, interesting subjects and nice people. Ten years that I’m looking for a good conference on AI in video games, I’m really happy to have found it.

Gabriel Robert

Senior AI programmer

Ubisoft, PARIS

Well done for putting on such a great conference, the program was varied, interesting and inspiring. Definitely hoping to return next year.

John Lusty

Lead Coder

Ninja Theory, CAMBRIDGE

The quality of the talks was consistently high and the event well organised. I thoroughly enjoyed the opportunity to discuss AI with so many people passionate about the subject and hope to return next year.

Vicky Smalley

Senior AI Programmer

Sony Computer Entertainment Europe

Thanks for organising a thoroughly enjoyable conference. It's nice to meet so many people with a common interest in AI. It was great to finally meet a number of people who I've spoken to via email over the past 2 years...

Neil Armstrong

Lead AI Programmer

Blitz Games Studios, ENGLAND

Thanks for hosting such an excellent conference. I found it great to meet such a broad cross section from the AI development community. It has certainly given me some inspiration for my upcoming projects. I will definitely try to come back next year!

David Partouche

Senior AI Engineer

CCP Games, ENGLAND

Congrats once again on pulling the conference off: it was by far the most inspiring work-related event I've ever been to. I hope it will become a regular event.

Remco Straatman

Lead AI Programmer

Guerrilla Games, AMSTERDAM



Day Tickets
(1 day)

All Access
(3 days)

Early Bird (limited)

All these tickets have gone already...

Sold Out!

Early Worm (timed)

Available in batches of 25 tickets until price increases.

from € 256

from € 540

Notes

  • The main amphitheatre (including some talks in this track) are accessible with regular Conference tickets, starting at €119.
  • Discounts are available for small teams and for those between jobs; email events(at)aigamedev.com for a coupon code.

Image Credit: Top 11 Football Manager

In this track, you'll learn about how and why dynamic pricing impacts revenues. Understand the special blend of skill and luck that goes into a casual game - and the extra ingredient that is needed to explain real player results. Discover effective computational approaches to complex balance problems, the challenges of real-time analytics, and the design of effective recommendation systems to power choices presented in game worlds. Thrill to the never before told story of how one studio detected and prevented early churners!

Applying analytics to game design is not about driving by looking in the rear view mirror while ranting about KPIs. Analytics can be used to apply real-time or just-in-time intelligent design judgement to the logic of who sees what when, and to the multidimensional dynamic challenges of balance and matchmaking. Our experts explain their creative analytical solutions to a variety of challenges in game design and experience design, across a variety of platforms and genres. Our goal is to give you some knowledge you can take home and apply. Come join the fun!

Keywords

  • machine learning
  • adaptive systems (pricing)
  • pattern recognition
  • real-time profiling

Format

  • Keynote
  • Presentations
  • Panel Discussion
  • Expert Q&A

Track Schedule (Day #2: Tuesday, July 21st)

Download Advanced Analytics & Data Science schedule in .ics format.

Track Schedule (Day #2: Tuesday, July 21st)

Day #2: Tuesday, July 21st

Procedural Content Generation

Agent Behavior

Advanced Analytics & Data Science

Bill Grosso, Scientific Revenue

11:00 Dynamic Pricing in Mobile Games (Amphitheatre)

Heather Stark and Bill Grosso

11:45 Extended Q&A on Dynamic Pricing and Conditional Offers (Laboratories)

Heather Stark, Kinran Limited

14:15 Analytics Standup Session and Open Q&A (Laboratories)

Download Advanced Analytics & Data Science schedule in .ics format.

Keynotes and Presentations

Dynamic Pricing in Mobile Games

Dynamic pricing techniques are at the forefront of adaptive systems in games. How can we better understand individual players and the player-base as a whole? What's necessary to design and tune a dynamic system that remains reliable even when operating without supervision? Which technologies are necessary to process the large quantities of data and extract meaningful models?

Originating in the airline industry almost 50 years ago, dynamic pricing has spread to the many other industries including retail, as well as In-App-Purchases within games or mobile applications. This talk will explain the challenges and investigate the opportunities: to reinvent how we look at data, and automate the process of sifting for information. We can take notions that are intuitively "obvious" to old-school marketers and automatically check whether the definitions actually capture actionable differences in consumer behavior.

Bill Grosso, Scientific Revenue

Bill is the founder and CEO of Scientific Revenue, the leading provider of dynamic pricing for in-app purchases. Prior to starting Scientific Revenue, Bill ran a consulting company focused on business intelligence and predictive analytics, and was an entrepreneur in residence at Rustic Canyon Venture Partners. Prior to those positions,...

Read full biography »

Bill Grosso, Scientific Revenue

Bill is the founder and CEO of Scientific Revenue, the leading provider of dynamic pricing for in-app purchases. Prior to starting Scientific Revenue, Bill ran a consulting company focused on business intelligence and predictive analytics, and was an entrepreneur in residence at Rustic Canyon Venture Partners. Prior to those positions,...

Read full biography »

Extended Q&A on Dynamic Pricing and Conditional Offers

Following up

Heather Stark, Kinran Limited

Heather is a games analyst who has worked on over a dozen games, as a consultant and in-house, across a wide range of genres and platforms. She is an active member of several London tech meetups and has a background as a consultant, tech analyst, and HCI researcher. Recently, she...

Read full biography »

Heather Stark, Kinran Limited

Heather is a games analyst who has worked on over a dozen games, as a consultant and in-house, across a wide range of genres and platforms. She is an active member of several London tech meetups and has a background as a consultant, tech analyst, and HCI researcher. Recently, she...

Read full biography »

Bill Grosso, Scientific Revenue

Bill is the founder and CEO of Scientific Revenue, the leading provider of dynamic pricing for in-app purchases. Prior to starting Scientific Revenue, Bill ran a consulting company focused on business intelligence and predictive analytics, and was an entrepreneur in residence at Rustic Canyon Venture Partners. Prior to those positions,...

Read full biography »

Bill Grosso, Scientific Revenue

Bill is the founder and CEO of Scientific Revenue, the leading provider of dynamic pricing for in-app purchases. Prior to starting Scientific Revenue, Bill ran a consulting company focused on business intelligence and predictive analytics, and was an entrepreneur in residence at Rustic Canyon Venture Partners. Prior to those positions,...

Read full biography »

Fast > Perfect: Practical Approximation Examples for Analytics using Spark Streaming

For mobile games, constant tweaks are the difference between success and failure. Game designers need metrics like DAU, new users and ARPDAU in real-time to be able to tweak quickly. But calculating, for example, uniqueness or newness of a data point requires a list of seen data points - both memory intensive and tricky when using real-time stream processing like Spark Streaming. Probabilistic data structures allow approximation of these properties with a fixed memory representation and are very well suited for stream processing. Getting from the theory of approximation to a practical useful metric at a low error rate even for many millions of users is another story. In this talk we will look at ways to achieve it:

  • Which approximation we use for selection of useful metrics
  • Why we picked a specific probabilistic data structure
  • How we store it in Cassandra as a time series
  • How we implemented it in Spark Streaming

Kevin Schmidt, Mind Candy

Kevin built up the data science and engineering team at Mind Candy and with the team created a scalable architecture for mobile game analytics based on Apache Spark. Before Mind Candy, Kevin was heading the data and back-end services team at Last.fm using Hadoop to work with ten years of...

Read full biography »

Kevin Schmidt, Mind Candy

Kevin built up the data science and engineering team at Mind Candy and with the team created a scalable architecture for mobile game analytics based on Apache Spark. Before Mind Candy, Kevin was heading the data and back-end services team at Last.fm using Hadoop to work with ten years of...

Read full biography »

Luis Angel Vicente Sanchez, Mind Candy

Luis is Senior Data Engineer at Mind Candy, was the first to introduce Spark Streaming at the company and is responsible for the real-time mobile analytics platform. He has more than 10 years of experience in software engineering and architecture, and has made contributions to open source projects like twitter/algebird,...

Read full biography »

Luis Angel Vicente Sanchez, Mind Candy

Luis is Senior Data Engineer at Mind Candy, was the first to introduce Spark Streaming at the company and is responsible for the real-time mobile analytics platform. He has more than 10 years of experience in software engineering and architecture, and has made contributions to open source projects like twitter/algebird,...

Read full biography »

Extending Recommendation Systems to Offer a Personalised in-game Experience

We will present the fundamentals and practical challenges of building a recommendation system for your online game, taking you through the established approaches of Collaborative Filtering, Content Based and Latent Factor based systems. After briefly discussing how to design, evaluate and deploy these algorithms we will demonstrate how recommendation systems can move beyond store front inventories to offer your players a personalised in-game experience. Currently deployed within RuneScape, the world’s largest free-to-play massively multiplayer online game, we will go behind the scenes and present the details of the "lucky challenge" system, an adaption of recommendation approaches to highlight personally relevant content to players exploring the world of Gielinor. Attempting to balance engagement, enjoyment and spend through a second-order set of real-time recommendations present unique challenges in terms of scalability, messaging, monitoring and algorithm design and our presentation will give a snapshot of our current solutions and future plans.

Simon Worgan, Jagex

Simon Worgan is the Lead Data Scientist at Jagex Games Studio. In this role he has applied his machine learning expertise to a variety of challenges within the games industry, these include in-game recommendation systems, sentiment analysis, behavioural player clustering and predictive modelling. With a Computer Science Ph.D. from the...

Read full biography »

Simon Worgan, Jagex

Simon Worgan is the Lead Data Scientist at Jagex Games Studio. In this role he has applied his machine learning expertise to a variety of challenges within the games industry, these include in-game recommendation systems, sentiment analysis, behavioural player clustering and predictive modelling. With a Computer Science Ph.D. from the...

Read full biography »

Early Churn Prediction and Personalised Interventions in TOP 11

How to tackle the problem of early churners, and define a strategy to overcome this issue. An approach to early churn prediction using machine learning algorithms is shared. Various machine learning models were tested, along with feature selection procedures. A clustering technique is introduced that segments users according to their first day gameplay habits. Also, a notification sending engine is described, which combines the knowledge gathered from the two previously mentioned methods, and targets the predicted churners with different types of personalized messages. This allows us to decrease early churn rate, while significantly increasing first day retention as well as boosting other metrics. We will also touch upon the underlying system infrastructure to deal with the scalability issues that occur when working with big data.

Miloš Milošević, Nordeus

Miloš is a data analyst at Nordeus, multinational social network game company, best known for Top Eleven, one of the most successful cross-platform sport games in the world. His main focus are predictive machine learning models and their deployment in a live production setting, as well as maintaining and further...

Read full biography »

Miloš Milošević, Nordeus

Miloš is a data analyst at Nordeus, multinational social network game company, best known for Top Eleven, one of the most successful cross-platform sport games in the world. His main focus are predictive machine learning models and their deployment in a live production setting, as well as maintaining and further...

Read full biography »

Measuring Player Skill at Complex Team Games in Fable Legends

When attempting to quantify player skill for segmentation or matchmaking within complex competitive games, it is crucial to find effective ways to describe a player’s skill in a given context. However, standard methods of measuring and inferring skill don’t work very well in this type of problem domain. We describe our approach to modelling player skill in the complex contexts that arise in Fable Legends, Lionhead’s upcoming action-RPG vs. RTS competitive title. Fable Legends has taken a number of innovative design decisions, from asymmetrical team sizes to dramatically varied player roles, varied enemy creature selections, abilities, items and map topologies. These design aspects raised some new challenges for the Legends data team, which has responsibility for providing quantitative support to game balancing and design. This talk discusses the challenges we faced and the learnings we gained from the modelling process. We initially found that modelling player skill using standard techniques proved to be time-consuming and ineffective. Recognising this, we dived deeper to improve the accuracy of our models,creatively building feature sets and moving from individual models to ensembles built using multiple techniques (including regression models, supervised learning and deep learning techniques). We also present an overview of applications of these techniques, with particular focus on how skill scoring enabled designers to target player skill levels (“pro gamers”, “casual gamers”) for balance and design changes.

John Hearty, Lionhead Studios

John works as a Data Scientist at Lionhead Studios, where he leads a cross-functional team in building a TB-scale analytics architecture. He also works closely with game design and product management stakeholders, providing modelling and analysis to support Fable Legends’ balance and design. He is currently authoring a book on...

Read full biography »

John Hearty, Lionhead Studios

John works as a Data Scientist at Lionhead Studios, where he leads a cross-functional team in building a TB-scale analytics architecture. He also works closely with game design and product management stakeholders, providing modelling and analysis to support Fable Legends’ balance and design. He is currently authoring a book on...

Read full biography »

Modeling Skill vs. Luck in Casual Game Levels

In casual games, players typically face an independent, concatenated set of game boards with different puzzles on them. These challenges are typically called levels. In characterising the differences between levels, the most common parameter used in the industry is the level’s success rate, an observational parameter that measures the probability of solving a level in one attempt.

This talk investigates additional empirically grounded ways to characterise differences between levels. We use both analytical characterisation of the game space, and empirical data from player experience in our investigation. We can characterise levels by considering how their play patterns are affected by the mixture of player skill and randomness. We propose (and provide model-based justication for two more parameters which can help us understand what is important about the design of a casual game level:

  • A way of describing the randomness of a level
  • A ‘discovery’ factor, related to the player’s exploration of the interaction possibilities within a level

The discovery factor initially appeared a residual of the analysis, but relates strongly to both design practice and theory.

Ivan Encinas, King

As a Senior Data Scientist, Ivan works on the Casual Game Performance Team, where provides analytical and model support to understand about the casual games at King, and their performance drivers.

At King he has worked with several game teams: Papa Pear, Diamond Digger and Bubble Witch 2. He...

Read full biography »

Ivan Encinas, King

As a Senior Data Scientist, Ivan works on the Casual Game Performance Team, where provides analytical and model support to understand about the casual games at King, and their performance drivers.

At King he has worked with several game teams: Papa Pear, Diamond Digger and Bubble Witch 2. He...

Read full biography »

Analytics Standup Session and Open Q&A

...

Heather Stark, Kinran Limited

Heather is a games analyst who has worked on over a dozen games, as a consultant and in-house, across a wide range of genres and platforms. She is an active member of several London tech meetups and has a background as a consultant, tech analyst, and HCI researcher. Recently, she...

Read full biography »

Heather Stark, Kinran Limited

Heather is a games analyst who has worked on over a dozen games, as a consultant and in-house, across a wide range of genres and platforms. She is an active member of several London tech meetups and has a background as a consultant, tech analyst, and HCI researcher. Recently, she...

Read full biography »

Speakers & Organizers

Bill Grosso

Heather Stark

Kevin Schmidt

Luis Angel Vicente Sanchez

Simon Worgan

Miloš Milošević

John Hearty

Ivan Encinas

Browse Conference Tracks

Character Animation Technology

Real-time Decisions

Agent Behavior

Procedural Content Generation

Virtual / Augmented Reality

Crowds and Ambient Life

Advanced Analytics & Data Science

Systemic Design, AI Directors

Cognitive Bots & Language