MACHINE LEARNING / DATA SCIENCE / DATA ENGINEERING
27th of November 2020
ONLINE
CONFERENCE STARTNS IN
  • 00

    days

  • 00

    hours

  • 00

    minutes

  • 00

    seconds

Add to My Calendar 2020-11-27 09:00:00 2020-11-27 17:00:00 Europe/Stockholm GAIA Conference 2020 For people with an interest in machine learning and data science. https://conference.gaia.fish Online GAIA conference@gaia.fish

ABOUT THE CONFERENCE

GAIA organises a one-day conference for people with interest in artificial intelligence and data science with the focus on what is going on within the field in Gothenburg. The aim is to create an environment for learning, networking, and knowledge-sharing among individuals, companies, organisations, and academia with a common interest. The conference focuses on applied machine learning and data science and introduces talks of diverse content given by enthusiastic people from the field, many with local connections.

Our last conference was live streamed on the 27th of November last year. We’re currently in progress of planning GAIA Conference 2021. Looking forward to seeing you there!

WHAT TO EXPECT
Inspiration and knowledge

Great talks will be given by representatives from academia and many different industries. We expect to get inspired and learn about different techniques, methods and tools used by engineers, data scientists and other people in the field. We hope to leave the conference with a long list of new things to learn more about and start exploring!

Food and drinks

Of course, food and drinks are included in the ticket price. We will provide you with breakfast, lunch and fika and infinite amounts of coffee and tea so that you can stay sharp the whole day. We recommend you plan for some extra time after the closing remarks as we are finishing the day with bubbles!

Location

After two years at Lindholmen, we are excited to host the conference in a more central location. This year we will be meeting at Svenska Mässan, which allows us to invite even more attendees, partner companies and startups and still have plenty of room to mingle!

One great conference

We’re honoured to have so many representatives from Gothenburg sharing their knowledge and thoughts. We’re looking forward to hearing about what is happening on the west coast and meeting other great talents with similar problems and interests.

SPEAKERS

We are thrilled about last year’s lineup! Check out the speakers and their abstracts to read more about the fantastic talks of 2020.

Zenodia Charpy

Senior Deep Learning Solution Architect | NVIDIA

Better, Faster utilizing your GPUs for deep learning workload

 

The utilization of GPUs has evolved from gaming purposes, to deep learning, and gear towards, but not limited to, computer vision type of workloads :autonomous vehicles, medical images in radiomics and video surveillance to name a few.

However, having powerful GPUs at your disposal is one thing, knowing how to use them to their full potential is something we are going to explore together.

Working at Nvidia, we are dedicated to create an eco-system with full cycle of hardware-and-software solutions, that enables optimization in all fronts: from providing plug-and-play-able docker repo maintained by us, to data augmentations within GPU on-the-fly, to parallel model training using multiple gpus with mixed_precision for target deployment.

We have open-sourced all these toolkits for data scientists to quickly kick-start and get to work, instead of spending precious time to fix environmental installation errors and bring super computing power at your fingertips.

Today we are going to take a look at these essential toolkits which enable deep learning practitioners, data scientists alike to improve their model training in parallel, iterate faster and develop better market-ready products.

 

 

Biography

Working many years hands-on as an in-house data scientist, an external deep-learning consultant , a cloud solution architect (on Azure) and now a senior deep learning solution architect at Nvidia. My journey of seeking the optimal pathways in utilizing multiple GPUs for deep learning is paved with years of industrial experiences and practical tips & tricks learned from pitfalls with real-world projects.I am on a mission to help data scientists and researchers alike to accelerate their deep learning workload with ease taking advantage of my learnings and experiences.

 

 

 

 

 

 

 

 

 

 

Thore Husfeldt

Professor of Computer Science | Lund University

An Introduction to Algorithmic Fairness

 

I will give a brief introduction to algorithmic fairness. I will introduce and compare several different definitions of fairness that can be used to evaluate outcomes of algorithmic decision procedures and highlight their relationship. Of particular interest are the tensions between notions of individual and group fairness, and between various notions of group fairness. The motivation and main take-aways from the presentation are aimed at a general academic audience, but I will include mathematically precise terms.

 

 

Biography

Thore Husfeldt is a Professor of Computer Science at Lund University and IT University of Copenhagen. He is a core researcher at the Basic Algorithms Research Copenhagen centre (BARC) and the current chair of Vetenskapsrådets expert panel for Computer Science. His research is in algebraic graph algorithms.

Robert Feldt

Professor of Software Engineering | Chalmers

Testing of Machine Learning Systems – the importance of “lagom” surprising inputs

 

Testing of machine learning (ML) components, such as deep neural nets, is not only about correctness and accuracy; there is a large number of quality properties that has to be ensured. While research on how to perform these different forms of testing is still immature it is growing at a tremendous pace. In this talk, I’ll give an overview of recent results on testing ML models and discuss how it differs from testing normal software. I will exemplify with recent work on how to find adequately (“lagom”) surprising test inputs, that are not random or look like noise, but rather that are realistic. While the current focus of research is mainly on neural nets I’ll also discuss if and how this might be generalised to other types of machine learning models.

Biography

Robert Feldt is a researcher and teacher with a passion for software and for augmenting humans with AI and Artificial Creativity/Innovation. He is a professor at Chalmers University of Technology in Gothenburg and frequently consults for companies in both Europe and Asia. He has broad interests spanning from human factors to hardcore automation and applied AI&statistics and works on software testing and quality, as well as human-centred (behavioural) software engineering.

Rocío Mercado

Postdoctoral Fellow | AstraZeneca

Discovering New Molecules Using Graph Neural Networks

 

There is growing interest in graph neural networks (GNNs) for graph representation learning. This is because graphs are natural data structures for describing an assortment of relational information, including molecular structures. One of the topics we are interested in within the Molecular AI (MAI) group at AstraZeneca is using GNNs for the design and discovery of new drug molecules.

The process of designing novel, drug-like molecules can be viewed as one of generating graphs which optimize all the features of the desirable molecules. We are interested in graph-based methods because they have the potential to capture a lot of molecular information with greater flexibility. Our generative models can quickly learn the underlying distribution of properties in training set molecules without any explicit writing of chemical rules.

In this talk, I will demonstrate how we can use deep learning methods such as GNNs to carry out pharmaceutical drug  discovery more efficiently.

Biography

I joined the Molecular AI group at AstraZeneca in October 2018. My work focuses on using deep learning methods for graph-based molecular design. Before AstraZeneca, I was a PhD student in Professor Berend Smit’s molecular simulation group at UC Berkeley and EPFL. I received my PhD in Chemistry from UC Berkeley in July 2018, and my BS in Chemistry from Caltech in June 2013.

Oscar Söderlund

Chief Software Architect | Einride

Scalable forecasting in Google Cloud

 

A practical and data engineering-oriented talk on how to stand up a scalable and fully automated pipeline for time series forecasting using Facebook’s Prophet library and the standard big data and machine learning tools in Google Cloud.

This will be done in the context of Einride’s data platform, which is all about creating actionable insights that drive customers toward sustainable transport. As a real-world case-study, I will show how we break down and understand transport demand in multiple dimensions – from a customer’s total demand down to thousands of individual sites and shipping lanes.

I will walk us through the full pipeline; extracting a production database dump, multiple tiers of data cleaning and transformation, how to use PySpark and Dataproc to parallelize model training and forecast generation, and how to orchestrate it all with Apache Airflow.

The key takeaway (and what’s really exciting) is how easy to use and available big data and machine learning tools have become – to tech giants and fledgling startups alike!

Biography

In 2018, Oscar left a cushy backend and data engineering job at Spotify to seek the thrill of a New Game+ experience from Gothenburg’s startup scene. Cue Einride, a crack team of technologists out to disrupt an outdated industry with sustainable transport solutions. Oscar has lately been working on building up Einride’s data platform capabilities and will be sharing practical advice on building scalable data pipelines from scratch in Google Cloud.

 

 

 

 

 

 

 

 

Pablo Estevez Castillo

Principal Data Scientist | Booking.com

What we have learned from 150 successful ML-enabled products at Booking.com

 

Booking.com is the world’s largest online travel agent where millions of guests find their accommodation and millions of accommodation providers list their properties including hotels, apartments, bed and breakfasts, guest houses, and more. During the last years we have applied Machine Learning to improve the experience of our customers and our business. While most of the Machine Learning literature focuses on the algorithmic or mathematical aspects of the field, not much has been published about how Machine Learning can deliver meaningful impact in an industrial environment where commercial gains are paramount. We conducted an analysis on about 150 successful customer facing applications of Machine Learning, developed by dozens of teams in Booking.com, exposed to hundreds of millions of users worldwide and validated through rigorous Randomized Controlled Trials. Following the phases of a Machine Learning project we describe our approach, the many challenges we found, and the lessons we learned while scaling up such a complex technology across our organization. Our main conclusion is that an iterative, hypothesis driven process, integrated with other disciplines was fundamental to build 150 successful products enabled by Machine Learning.

Biography

Pablo Estevez is Principal Data Scientist at Booking.com. He has worked on recommendations, personalisation and experimentation accross Booking.com website, as well as a manager on several machine learning, data science and product development teams.

Olof Mogren

Machine Learning researcher | RISE

Social bias and fairness in natural language processing

 

Learned continuous representations for language units was the first trembling steps of making neural networks useful for natural language processing (NLP), and promised a future with semantically rich representations for downstream solutions. NLP has now seen some of the progress that previously happened in image processing: the availability of increased computing power and the development of algorithms have allowed people to train larger models that perform better than ever. Such models also make it possible to use transfer learning for language tasks, thus leveraging large widely available datasets.

In 2016, Bolukbasi, et.al., presented their paper “Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings”, shedding lights on some of the gender bias that was available in trained word embeddings at the time. Datasets obviously encode the social bias that surrounds us, and models trained on that data may expose the bias in their decisions. It is important to be aware of what information a learned system is basing its predictions on. Some solutions have been proposed to limit the expression of societal bias in NLP systems. These include techniques such as data augmentation and representation calibration. Similar approaches may also be relevant for privacy and disentangled representations. In this talk, we’ll discuss some of these issues, and go through some of the solutions that have been proposed recently.

Biography

Olof Mogren is heading the AI research group at RISE in Gothenburg. Research interests include representation learning, data privacy, and modelling the world around us, in application domains such as analysis of medical texts, image processing, and sensor modelling.

Rebecka Jacobsson

Machine Learning Lead | iZettle

Building a feature library for machine learning at iZettle

 

When we started building features for our first machine learning models at iZettle, we quickly realized we didn’t want to reinvent the wheel for every model we built. Instead, we created a shared feature building framework, which over time has evolved into a core part of our machine learning infrastructure. In this talk, I will explain why you might want to create a feature library, suggest some important trade-offs to consider and share how we first built the simplest solution possible and then iterated from there.

Biography

Rebecka is the Machine Learning Lead at iZettle, a member of the PayPal family, that is on a mission to help small businesses succeed in a world of giants. She manages a team of data scientists and machine learning engineers working on problems ranging from fraud detection and credit scoring, to marketing optimization and lead generation. Rebecka is also one of the founders of Women in Data Science Sweden, a network for current and aspiring female data scientists that arranges technical conferences and events.

 

 

 

 

 

 

 

 

 

Daniel Gillblad

Head of AI Research | RISE

AI – Perspectives and Strategies

 

While AI and its applications continue to develop rapidly, many challenges remain to realise the full potential. We will reflect over the developments within AI and Machine Learning during the last few years, the current state-of-the-art, what remains to be done and what some of the urgent issues are at the moment. With this as a starting point, we will discuss the current state of AI and AI applications in Sweden and how we can think about national strategies for accelerating the use of AI nationally, in both industry and public sector. Finally, we will provide some examples of ongoing work in important areas.

Biography

Daniel Gillblad is the Head of AI Research at RISE, the Swedish Research Institutes, and Co-Director of AI-Innovation of Sweden. He has a background in machine learning and large scale data analytics, and has extensive experience of applying such methods in practice.

Christian Forssén

Professor in Theoretical Physics | Chalmers

Learning from data in fundamental physics research

 

Bayesian statistics allows to quantify the strength of inductive inference from facts (such as experimental data) to propositions such as scientific hypotheses and models. It can be used for example in searches for physics beyond the standard model, and in multi-message astronomy for the analysis of gravitational wave signals. In our theoretical-physics research we develop novel machine-learning methods to make the Bayesian approach tractable for models that involve heavy computing.

A general message of this presentation will be that the Bayesian philosophy can be embraced by machine-learning practitioners to introduce uncertainties and to avoid overfitting.

Biography

Christian Forssén is Professor in theoretical physics at Chalmers. He uses high-performance computing, Bayesian statistics and machine learning, to address basic-science questions within theoretical nuclear and particle physics. Christian has supervised research projects for more than 100 bachelor and master students and has been teaching both undergraduate and graduate-level physics classes, as well as advanced courses on Bayesian data analysis and the use of machine learning methods in physics.

Daniel Langkilde

Co-founder | Annotell

The impact of data quality on model performance

 

A lot of focus in the machine learning community is on the choice of algorithm. This talk will instead focus on the dataset. While most people by now have realized that dataset quality is critical for the success of machine learning projects, most teams still struggle to quantify data quality. This talk will provide hands-on examples of how to quantify dataset quality, and also reason about the impact various types of quality issues can be expected to have on model performance.

 

Biography

Daniel Langkilde is CEO and co-founder of Annotell, which provides the analytics and annotation platform used to ensure the performance of autonomous vehicle perception systems. He has focused for almost 10 years on the relationship between data quality and machine learning product performance. He has an M.Sc. in Engineering Mathematics and has been a Visiting Scholar at both UC Berkeley and MIT. Before starting Annotell he was Team Lead for Collection & Analysis at Recorded Future. Besides that, he is also on the Board of Directors at Chalmers University.

Martin Tegner

Machine learning researcher | IKEA Digital

Data privacy and fairness in recommender systems

 

In early 2020, IKEA made a promise to give back the control to customers over their data. In this talk, we look at two aspects of the promise in the context of data-driven product recommendations. First, we analyse performance of such systems under minimal data requirements. Second, we propose a Bayesian approach to recommendations that uses in-session active learning to give personalised content without collection of private data.

 

Biography

Martin is a data science researcher at IKEA group. He works with data-powered AI to enrich the customer experience at every point of the customer journey. In his current research, he develops algorithms that provide personalised content while respecting fair use of the customer’s data. Prior to joining IKEA, he was a researcher in the machine learning group at the University of Oxford.

Adam Andersson

Radar Systems Engineer | Saab Surveillance

Removing computational bottlenecks with deep learning

 

Deep learning has transformed the fields of computer vision and natural language processing. There is a more recent trend in using neural networks to solve differential equations. It is less well known and has so far mostly gained academic interest. Computer aided engineering is, under the hood, based on solving partial differential equations (PDE) approximately with classical methods such as the finite element method.

A crash simulation takes hours to run on a computer cluster and the PDE has 4 dimensions (three space dimensions and one time dimension). Last year, various nonlinear PDE with 10000 space variables were solved approximately with deep learning on much smaller computers. The problems are not perfectly comparable but this result still indicates the ability of deep learning to scale beyond classical methods. Both automatic control and sensor fusion problems can be formulated and solved in terms of PDE, but due to computational complexity of classical methods these are not used in practical algorithms. In this talk I discuss how such equations for stochastic control and potentially for sensor fusion can be used for deriving practical algorithms, with deep learning being the enabler.

Biography

Adam Andersson works as radar systems engineer at Saab Surveillance. He is a mathematician with research background in numerical analysis and probability theory. During his four years in industry he has been working with applications of machine learning and sensor fusion, as well as more research oriented machine learning projects. His role has a semi-academic character with collaborations, master and PhD supervision at Chalmers.

Carl Lindberg

Co-Founder / Director of Strategic Data Applications | Sigmastocks / Zenseact

Common Pitfalls when Working with Data

 

With the present AI and Data Science hype, the marvelous theory of Statistics has found yet two more fields to be applied to. However, the Statistical Machine works differently than most other machines you have heard of, and there are many pitfalls into which Data Scientists innocent of Applied Statistics too easily fall. This talk will deal with some of the most common such pitfalls, illustrated by relevant real life examples.

 

 

 

Biography

Carl is an entrepreneur who has co-founded e g Sigmastocks and Islands of Math. He also has a background as professor in Mathematical Statistics. Currently, he coaches PhD students at Zenseact, a Swedish autonomous driving company.

SPEAKERS

We are thrilled about last year’s lineup! Check out the speakers and their abstracts to read more about the fantastic talks of 2020.

Zenodia Charpy

Senior Deep Learning Solution Architect | NVIDIA

Better, Faster utilizing your GPUs for deep learning workload

 

The utilization of GPUs has evolved from gaming purposes, to deep learning, and gear towards, but not limited to, computer vision type of workloads :autonomous vehicles, medical images in radiomics and video surveillance to name a few.

However, having powerful GPUs at your disposal is one thing, knowing how to use them to their full potential is something we are going to explore together.

Working at Nvidia, we are dedicated to create an eco-system with full cycle of hardware-and-software solutions, that enables optimization in all fronts: from providing plug-and-play-able docker repo maintained by us, to data augmentations within GPU on-the-fly, to parallel model training using multiple gpus with mixed_precision for target deployment.

We have open-sourced all these toolkits for data scientists to quickly kick-start and get to work, instead of spending precious time to fix environmental installation errors and bring super computing power at your fingertips.

Today we are going to take a look at these essential toolkits which enable deep learning practitioners, data scientists alike to improve their model training in parallel, iterate faster and develop better market-ready products.

 

 

Biography

Working many years hands-on as an in-house data scientist, an external deep-learning consultant , a cloud solution architect (on Azure) and now a senior deep learning solution architect at Nvidia. My journey of seeking the optimal pathways in utilizing multiple GPUs for deep learning is paved with years of industrial experiences and practical tips & tricks learned from pitfalls with real-world projects.I am on a mission to help data scientists and researchers alike to accelerate their deep learning workload with ease taking advantage of my learnings and experiences.

 

Oscar Söderlund

Chief Software Architect | Einride

Scalable forecasting in Google Cloud

 

A practical and data engineering-oriented talk on how to stand up a scalable and fully automated pipeline for time series forecasting using Facebook’s Prophet library and the standard big data and machine learning tools in Google Cloud.

This will be done in the context of Einride’s data platform, which is all about creating actionable insights that drive customers toward sustainable transport. As a real-world case-study, I will show how we break down and understand transport demand in multiple dimensions – from a customer’s total demand down to thousands of individual sites and shipping lanes.

I will walk us through the full pipeline; extracting a production database dump, multiple tiers of data cleaning and transformation, how to use PySpark and Dataproc to parallelize model training and forecast generation, and how to orchestrate it all with Apache Airflow.

The key takeaway (and what’s really exciting) is how easy to use and available big data and machine learning tools have become – to tech giants and fledgling startups alike!

Biography

In 2018, Oscar left a cushy backend and data engineering job at Spotify to seek the thrill of a New Game+ experience from Gothenburg’s startup scene. Cue Einride, a crack team of technologists out to disrupt an outdated industry with sustainable transport solutions. Oscar has lately been working on building up Einride’s data platform capabilities and will be sharing practical advice on building scalable data pipelines from scratch in Google Cloud.

Daniel Langkilde

Co-founder | Annotell

The impact of data quality on model performance

 

A lot of focus in the machine learning community is on the choice of algorithm. This talk will instead focus on the dataset. While most people by now have realized that dataset quality is critical for the success of machine learning projects, most teams still struggle to quantify data quality. This talk will provide hands-on examples of how to quantify dataset quality, and also reason about the impact various types of quality issues can be expected to have on model performance.

 

Biography

Daniel Langkilde is CEO and co-founder of Annotell, which provides the analytics and annotation platform used to ensure the performance of autonomous vehicle perception systems. He has focused for almost 10 years on the relationship between data quality and machine learning product performance. He has an M.Sc. in Engineering Mathematics and has been a Visiting Scholar at both UC Berkeley and MIT. Before starting Annotell he was Team Lead for Collection & Analysis at Recorded Future. Besides that, he is also on the Board of Directors at Chalmers University.

Carl Lindberg

Co-Founder / Director of Strategic Data Applications | Sigmastocks / Zenseact

Common Pitfalls when Working with Data

 

With the present AI and Data Science hype, the marvelous theory of Statistics has found yet two more fields to be applied to. However, the Statistical Machine works differently than most other machines you have heard of, and there are many pitfalls into which Data Scientists innocent of Applied Statistics too easily fall. This talk will deal with some of the most common such pitfalls, illustrated by relevant real life examples.

 

 

 

Biography

Carl is an entrepreneur who has co-founded e g Sigmastocks and Islands of Math. He also has a background as professor in Mathematical Statistics. Currently, he coaches PhD students at Zenseact, a Swedish autonomous driving company.

Rebecka Jacobsson

Machine Learning Lead | iZettle

Building a feature library for machine learning at iZettle

 

When we started building features for our first machine learning models at iZettle, we quickly realized we didn’t want to reinvent the wheel for every model we built. Instead, we created a shared feature building framework, which over time has evolved into a core part of our machine learning infrastructure. In this talk, I will explain why you might want to create a feature library, suggest some important trade-offs to consider and share how we first built the simplest solution possible and then iterated from there.

Biography

Rebecka is the Machine Learning Lead at iZettle, a member of the PayPal family, that is on a mission to help small businesses succeed in a world of giants. She manages a team of data scientists and machine learning engineers working on problems ranging from fraud detection and credit scoring, to marketing optimization and lead generation. Rebecka is also one of the founders of Women in Data Science Sweden, a network for current and aspiring female data scientists that arranges technical conferences and events.

Olof Mogren

Machine Learning researcher | RISE

Social bias and fairness in natural language processing

 

Learned continuous representations for language units was the first trembling steps of making neural networks useful for natural language processing (NLP), and promised a future with semantically rich representations for downstream solutions. NLP has now seen some of the progress that previously happened in image processing: the availability of increased computing power and the development of algorithms have allowed people to train larger models that perform better than ever. Such models also make it possible to use transfer learning for language tasks, thus leveraging large widely available datasets.

In 2016, Bolukbasi, et.al., presented their paper “Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings”, shedding lights on some of the gender bias that was available in trained word embeddings at the time. Datasets obviously encode the social bias that surrounds us, and models trained on that data may expose the bias in their decisions. It is important to be aware of what information a learned system is basing its predictions on. Some solutions have been proposed to limit the expression of societal bias in NLP systems. These include techniques such as data augmentation and representation calibration. Similar approaches may also be relevant for privacy and disentangled representations. In this talk, we’ll discuss some of these issues, and go through some of the solutions that have been proposed recently.

Biography

Olof Mogren is heading the AI research group at RISE in Gothenburg. Research interests include representation learning, data privacy, and modelling the world around us, in application domains such as analysis of medical texts, image processing, and sensor modelling.

Rocío Mercado

Postdoctoral Fellow | AstraZeneca

Discovering New Molecules Using Graph Neural Networks

 

There is growing interest in graph neural networks (GNNs) for graph representation learning. This is because graphs are natural data structures for describing an assortment of relational information, including molecular structures. One of the topics we are interested in within the Molecular AI (MAI) group at AstraZeneca is using GNNs for the design and discovery of new drug molecules.

The process of designing novel, drug-like molecules can be viewed as one of generating graphs which optimize all the features of the desirable molecules. We are interested in graph-based methods because they have the potential to capture a lot of molecular information with greater flexibility. Our generative models can quickly learn the underlying distribution of properties in training set molecules without any explicit writing of chemical rules.

In this talk, I will demonstrate how we can use deep learning methods such as GNNs to carry out pharmaceutical drug  discovery more efficiently.

Biography

I joined the Molecular AI group at AstraZeneca in October 2018. My work focuses on using deep learning methods for graph-based molecular design. Before AstraZeneca, I was a PhD student in Professor Berend Smit’s molecular simulation group at UC Berkeley and EPFL. I received my PhD in Chemistry from UC Berkeley in July 2018, and my BS in Chemistry from Caltech in June 2013.

Adam Andersson

Radar Systems Engineer | Saab Surveillance

Removing computational bottlenecks with deep learning

 

Deep learning has transformed the fields of computer vision and natural language processing. There is a more recent trend in using neural networks to solve differential equations. It is less well known and has so far mostly gained academic interest. Computer aided engineering is, under the hood, based on solving partial differential equations (PDE) approximately with classical methods such as the finite element method.

A crash simulation takes hours to run on a computer cluster and the PDE has 4 dimensions (three space dimensions and one time dimension). Last year, various nonlinear PDE with 10000 space variables were solved approximately with deep learning on much smaller computers. The problems are not perfectly comparable but this result still indicates the ability of deep learning to scale beyond classical methods. Both automatic control and sensor fusion problems can be formulated and solved in terms of PDE, but due to computational complexity of classical methods these are not used in practical algorithms. In this talk I discuss how such equations for stochastic control and potentially for sensor fusion can be used for deriving practical algorithms, with deep learning being the enabler.

Biography

Adam Andersson works as radar systems engineer at Saab Surveillance. He is a mathematician with research background in numerical analysis and probability theory. During his four years in industry he has been working with applications of machine learning and sensor fusion, as well as more research oriented machine learning projects. His role has a semi-academic character with collaborations, master and PhD supervision at Chalmers.

Christian Forssén

Professor in Theoretical Physics | Chalmers

Learning from data in fundamental physics research

 

Bayesian statistics allows to quantify the strength of inductive inference from facts (such as experimental data) to propositions such as scientific hypotheses and models. It can be used for example in searches for physics beyond the standard model, and in multi-message astronomy for the analysis of gravitational wave signals. In our theoretical-physics research we develop novel machine-learning methods to make the Bayesian approach tractable for models that involve heavy computing.

A general message of this presentation will be that the Bayesian philosophy can be embraced by machine-learning practitioners to introduce uncertainties and to avoid overfitting.

Biography

Christian Forssén is Professor in theoretical physics at Chalmers. He uses high-performance computing, Bayesian statistics and machine learning, to address basic-science questions within theoretical nuclear and particle physics. Christian has supervised research projects for more than 100 bachelor and master students and has been teaching both undergraduate and graduate-level physics classes, as well as advanced courses on Bayesian data analysis and the use of machine learning methods in physics.

Robert Feldt

Professor of Software Engineering | Chalmers

Testing of Machine Learning Systems – the importance of “lagom” surprising inputs

 

Testing of machine learning (ML) components, such as deep neural nets, is not only about correctness and accuracy; there is a large number of quality properties that has to be ensured. While research on how to perform these different forms of testing is still immature it is growing at a tremendous pace. In this talk, I’ll give an overview of recent results on testing ML models and discuss how it differs from testing normal software. I will exemplify with recent work on how to find adequately (“lagom”) surprising test inputs, that are not random or look like noise, but rather that are realistic. While the current focus of research is mainly on neural nets I’ll also discuss if and how this might be generalised to other types of machine learning models.

Biography

Robert Feldt is a researcher and teacher with a passion for software and for augmenting humans with AI and Artificial Creativity/Innovation. He is a professor at Chalmers University of Technology in Gothenburg and frequently consults for companies in both Europe and Asia. He has broad interests spanning from human factors to hardcore automation and applied AI&statistics and works on software testing and quality, as well as human-centred (behavioural) software engineering.

 

 

 

 

 

 

 

 

 

Thore Husfeldt

Professor of Computer Science | Lund University

An Introduction to Algorithmic Fairness

 

I will give a brief introduction to algorithmic fairness. I will introduce and compare several different definitions of fairness that can be used to evaluate outcomes of algorithmic decision procedures and highlight their relationship. Of particular interest are the tensions between notions of individual and group fairness, and between various notions of group fairness. The motivation and main take-aways from the presentation are aimed at a general academic audience, but I will include mathematically precise terms.

 

 

Biography

Thore Husfeldt is a Professor of Computer Science at Lund University and IT University of Copenhagen. He is a core researcher at the Basic Algorithms Research Copenhagen centre (BARC) and the current chair of Vetenskapsrådets expert panel for Computer Science. His research is in algebraic graph algorithms.

 

 

 

 

 

 

 

 

 

Daniel Gillblad

Head of AI Research | RISE

AI – Perspectives and Strategies

 

While AI and its applications continue to develop rapidly, many challenges remain to realise the full potential. We will reflect over the developments within AI and Machine Learning during the last few years, the current state-of-the-art, what remains to be done and what some of the urgent issues are at the moment. With this as a starting point, we will discuss the current state of AI and AI applications in Sweden and how we can think about national strategies for accelerating the use of AI nationally, in both industry and public sector. Finally, we will provide some examples of ongoing work in important areas.

Biography

Daniel Gillblad is the Head of AI Research at RISE, the Swedish Research Institutes, and Co-Director of AI-Innovation of Sweden. He has a background in machine learning and large scale data analytics, and has extensive experience of applying such methods in practice.

 

 

 

 

 

 

 

 

Pablo Estevez Castillo

Principal Data Scientist | Booking.com

What we have learned from 150 successful ML-enabled products at Booking.com

 

Booking.com is the world’s largest online travel agent where millions of guests find their accommodation and millions of accommodation providers list their properties including hotels, apartments, bed and breakfasts, guest houses, and more. During the last years we have applied Machine Learning to improve the experience of our customers and our business. While most of the Machine Learning literature focuses on the algorithmic or mathematical aspects of the field, not much has been published about how Machine Learning can deliver meaningful impact in an industrial environment where commercial gains are paramount. We conducted an analysis on about 150 successful customer facing applications of Machine Learning, developed by dozens of teams in Booking.com, exposed to hundreds of millions of users worldwide and validated through rigorous Randomized Controlled Trials. Following the phases of a Machine Learning project we describe our approach, the many challenges we found, and the lessons we learned while scaling up such a complex technology across our organization. Our main conclusion is that an iterative, hypothesis driven process, integrated with other disciplines was fundamental to build 150 successful products enabled by Machine Learning.

Biography

Pablo Estevez is Principal Data Scientist at Booking.com. He has worked on recommendations, personalisation and experimentation accross Booking.com website, as well as a manager on several machine learning, data science and product development teams.

Martin Tegner

Machine learning researcher | IKEA Digital

Data privacy and fairness in recommender systems

 

In early 2020, IKEA made a promise to give back the control to customers over their data. In this talk, we look at two aspects of the promise in the context of data-driven product recommendations. First, we analyse performance of such systems under minimal data requirements. Second, we propose a Bayesian approach to recommendations that uses in-session active learning to give personalised content without collection of private data.

 

Biography

Martin is a data science researcher at IKEA group. He works with data-powered AI to enrich the customer experience at every point of the customer journey. In his current research, he develops algorithms that provide personalised content while respecting fair use of the customer’s data. Prior to joining IKEA, he was a researcher in the machine learning group at the University of Oxford.

SCHEDULE

The conference was live streamed during the whole day of November 27th!  We packed the day full of interesting talks. To read more about the individual talks, please check out the speaker profiles. Make sure to check out the ones of extra interest to you on our youtube channel.

Opening Remarks

by Jakob Andersson

Chairman | GAIA

09:00

Keynote: An Introduction to Algorithmic Fairness

by Thore Husfeldt

Professor of Computer Science | Lund University

09:15

Social bias and fairness in natural language processing

by Olof Mogren

Machine Learning researcher | RISE

10:00

Break

10:20

Common Pitfalls when Working with Data

by Carl Lindberg

Co-Founder Sigmastocks and Director of Strategic Data Applications at Zenseact

10:35

Data privacy and fairness in recommender systems

by Martin Tegner

Machine learning researcher | IKEA Digital

10:55

Building a feature library for machine learning at iZettle

by Rebecka Jacobsson

Machine Learning Lead | iZettle

11:20

Scalable forecasting in Google Cloud

by Oscar Söderlund

Chief Software Architect | Einride

11.40

Lunch Break

12:00

Intro

13:00

Keynote: What we have learned from 150 successful ML-enabled products at Booking.com

by Pablo Estevez Castillo

Principal Data Scientist | Booking.com

13:05

The impact of data quality on model performance

by Daniel Langkilde

Co-founder | Annotell

13:45

Testing of Machine Learning Systems – the importance of “lagom” surprising inputs

by Robert Feldt

Professor of Software Engineering | Chalmers

14:05

Break

14:25

Learning from data in fundamental physics research

by Christian Forssén

Professor in Theoretical Physics | Chalmers

14:35

Discovering New Molecules Using Graph Neural Networks

by Rocío Mercado

Postdoctoral Fellow | AstraZeneca

14:55

Removing computational bottlenecks with deep learning

by Adam Andersson

Radar Systems Engineer | Saab Surveillance

15:15

Better, Faster utilizing your GPUs for deep learning workload

by Zenodia Charpy

Senior Deep Learning Solution Architect ( Automotive / health ) | Nvidia

15:35

Break

15:55

Keynote: AI – Perspectives and Strategies

by Daniel Gillblad

RISE/Recorded Future

16:05

Closing Remarks

by Josef Lindman Hörnlund

Board member | GAIA

16:50

OUR PARTNERS

It is because of these fantastic companies that we were able to arrange the GAIA Conference! Reach out to us if you are interested in becoming a partner of GAIA Conference 2021. More information about next year’s conference will be available during spring.

OUR PARTNERS

It is because of these fantastic companies that we were able to arrange the GAIA Conference! Reach out to us if you are interested in becoming a partner of GAIA Conference 2021. More information about next year’s conference will be available during spring.

CONTACT

Program and speakers

Are you wondering about the schedule or any of our sessions? Send an email to program@gaia.fish.

Questions or comments

Do you have any other questions or comments to us, send an email to conf@gaia.fish and we’ll promise to help you!

Partnerships

Want to support the conference by becoming a partner? Reach out to us on partnership@gaia.fish for more information.

© 2021 GAIA

conference@gaia.fish

© 2021 GAIA

conference@gaia.fish