Track / Overview

Geosciences & Remote Sensing

There is an exponential increase in the volume and variety of geo- and environmental data coming from different sources, including numerous earth observation monitoring networks and remote sensing images. In order to be used in intelligent decision making process, these data and information have to leverage machine learning in order to be efficiently processed and understood. In this session we will consider the application of Machine Learning, which is nonlinear, robust and universal tool for the analysis, modelling and visualisation of complex phenomena, in geoscience & remote sensing fundamental and applied studies for the recognition and classification of environmental patterns.


Maintaining and protecting biodiversity of plants and animals is essential for our life on earth. Due to climate change, human destruction of precious habitats, and industrial agriculture, biodiversity is decreasing in many countries. A major bottleneck for countermeasures is lack of accurate, dense biodiversity data at large scale. Today, biodiversity is mainly measured with field surveys that assess biodiversity manually in situ, which is labor-intensive, costly, and delivers only scarce, point-wise data with long revisit cycles. In this session, we will explore how machine learning can help automating, scaling, and improving quality of biodiversity estimation to help protecting the environment.

Sustainability & Energy

The exponential growth of various types of data together with the availability of machine learning and artificial intelligence methods are dramatically impacting the urban energy sector. It generates big data through smart meters, sensor networks, customer payments, satellite imagery, etc. This session provides a framework to discuss the use of big data and associate analytic methods in the urban energy sector with emphasis on machine learning and artificial intelligence methods for modelling and optimization of power generation and heat production. Furthermore, the session explores means of recognizing patterns in energy consumption as well as forecasting the energy resource potentials.


Droughts, heat waves, floods and storms induced by climate changes are some of the major natural disasters which are directly impacting population and society. Researchers develop predictive tools that can possibly reduce adverse impacts by allowing some kind of preventive action. Using artificial intelligence on the flood of data that is generated every day from sensors, gauges and monitors will allow to spot patterns quickly and automatically. In this session, we will explore how machine learning can help improving climate forecasts, better identifying atmospheric processes for building a resilient framework to face the effects of climate change.

Track / Schedule

Semantic interpretation of optical remote sensing data by computer vision and machine learning

With Michele Volpi

Machine Learning in solving subsurface energy resources exploration and development challenges: Discover, Describe, Predict, Decide

With Vasily Demyanov

Current application and expectations of Machine Learning for the Agriculture finance industry

With Sylvain Coutu

Artificial Environmental Intelligence With High Flying, Far Walking And Deep Learning

With Sylvain Coutu

Deep Learning for Land Use/Cover Statistics of Switzerland

With Maria Schönholzer

Detection of shallow landslides on aerial images using convolutional neural networks

With Maxim Samarin

Coffee Break

Novel technologies, data and methods to predict and manage global biodiversity change

With Walter Jetz

iNaturalist: Large Scale Visual Classification of the Natural World

With Grant van Horn

Panel Discussion

With Vasily Demyanov, Michele Volpi, Grant van Horn & Walter Jetz

ML & Environmental Risks

With Mikhail Kanevski

Learning and Optimization for Environment

With Saman Halgamuge

Machine learning for sustainability assessment from a Life Cycle Assessment perspective

With Antonino Marvuglia

How to estimate the electricity potential of roof mounted PV panels in a country, with a little bit of data

With Dan Assouline

AI enabled biofouling monitoring and cleaning system for offshore wind turbine monopile foundations

With Bojie Sheng

Coffee Break

Machine learning and snowflakes

With Alexis Berne

Climate change detection: A case for applied statistical learning?

With Sebastian Sippel

Panel Discussion

With Alexis Berne, Sebastian Sippel & Saman Halgamuge

Track / Speakers

Mikhail Kanevski

Professor, UNIL

Vasily Demyanov

Professor, Heriot-Watt University Edinburgh

Sylvain Coutu

Senior Agro Underwriter, product and R&D manager, Swiss Re

Maria Schönholzer

Scientific Assistant, FHNW

Walter Jetz

Professor, Yale University

Grant van Horn

Dr, Caltech

Saman Halgamuge

Professor, Australian National University

Antonino Marvuglia

Senior R&T Associate, Luxembourg Institute of Science and Technology (LIST)

Dan Assouline

PhD Student, EPFL

Bojie Sheng

Research Fellow, Brunel University London

Sebastian Sippel

PostDoc, ETH Zürich

Alexis Berne

Professor, EPFL

Michele Volpi

Senior Data Scientist, Swiss Data Science Center (ETH Zürich/EPFL)

Maxim Samarin

PhD Student, University of Basel

Track / Co-organizers

Jan Dirk Wegner

Head, EcoVision Lab, ETH Zurich

Alina Walch

PhD Student, EPFL

Roberto Castello

Scientific Collaborator, EPFL

Mikhail Kanevski

Professor, UNIL

Nahid Mohajeri

PostDoc, Oxford University

Frank de Morsier

Chief Technology Officer, Picterra

AMLD EPFL 2019 / Tracks & talks

AI & Environment

Mikhail Kanevski, Vasily Demyanov, Sylvain Coutu, Maria Schönholzer, Walter Jetz, Grant van Horn, Saman Halgamuge, Antonino Marvuglia, Dan Assouline, Bojie Sheng, Sebastian Sippel, Alexis Berne, Michele Volpi, Maxim Samarin

13:30-17:00 January 2809:00-12:30 January 293BC

AI & Cities

Mohamed Kafsi, Cristina Kadar, Alex “Sandy” Pentland, Valentine Goddard, Daniel Gatica-Perez, Piotr Mirowski, Niklas Goby, Christopher Nowzohour, Stephen Goldsmith

13:30-17:00 January 284BC

AI & Computer Systems

Kevin Smeyers, Michael Papamichael, Andreas Moshovos, Hadi Esmaeilzadeh, Svetlana Levitan

09:00-12:30 January 294BC

AMLD / Global partners