Department of Computer Science
(on leave from IIT Delhi)
Subhashis Banerjee is a professor in the Department of Computer Science at Ashoka University. He is on leave from the Department of Computer Science and Engineering at IIT Delhi, where he has held the Ministry of Urban Development, Microsoft and Naren Gupta chair professorships. Subhashis is also associated with the School of Public Policy and the Centre for Transportation Research at IIT Delhi.
Subhashis' primary areas of research are computer vision and machine learning, with a special emphasis on geometric algorithms. He has worked on problems related to structure from motion, large scale 3D reconstruction, SLAM, Graph Cut methods in computer vision, autonomous driving and ADAS, machine learning, prediction of socio-economic development indicators from satellite images and computational radiology. He has been on the editorial boards of the International journal of Computer Vision and Computers and Graphics, and has published in leading journals and conferences.
Recently he has also developed an interest in policy issues in digital identity, digitisation and society, electronic voting, data and privacy protection, and fairness and reliability of machine learning algorithms.
Subhashis graduated in electrical Engineering from Jadavpur University in 1982 and did his Master's and PhD from the Indian Institute of Science in 1984 and 1989 respectively.
Abstract of Proposed Discussion:
Subhashis Banerjee will be co-chairing the session on AI and sustainability which will cover issues related to climate and environment, plant biology and sustainable agriculture, ecology, forest degradation and forest use patterns.
Hessian Center for Artificial Intelligence
My team and I would like to make computers learn so much about the world, so rapidly and flexibly, as humans. This poses many deep and fascinating scientific problems: How can computers learn with less help from us and data? How can computers reason about and learn with complex data such as graphs and uncertain databases? How can pre-existing knowledge be exploited? How can computers decide autonomously which representation is best for the data at hand? Can learned results be physically plausible or be made understandable by us? How can computers learn together with us in the loop? To this end, my team and I develop novel machine learning (ML) and artificial intelligence (AI) methods, i.e., novel computational methods that contain and combine, for example search, logical and probabilistic techniques as well as (deep) (un)supervised and reinforcement learning methods. Currently, we focus specifically on probabilistic circuits, causality, explanatory interactive learning, probabilistic programming, and ethics in AI in order to push the third wave of AI.
Most of AI in use today falls under the categories of the first two waves of AI research. First wave AIs follow clear rules, written by people, aiming to cover every eventuality. Second wave AIs are the kind that use statistical learning to arrive at an answer for a certain type of problem—think of image recognition systems. The Third wave of AI envisions a future in which machines are more than just tools that execute human programmed rules or generalize from human-curated data sets. Rather, the machines it envisions will function more as colleagues than as tools. AI systems of this third wave of AI can acquire human-like communication and reasoning capabilities, with the ability to recognize new situations and to adapt to them. For example, a third wave AI might note that a speed limit of 120 km/h does not make sense when entering a small village.
Centre of Excellence in Artificial Intelligence, IIT Kharagpur
My research interests are related to Machine Learning, Data Science and their applications in domains such as earth sciences, especially climate. My long-term research goal is to contribute to the Sustainable Development Goals of the United Nations Development Program using these approaches. I am interested in developing models of complex physical processes and discovering scientific insights from them using data. I am particularly interested in understanding the vagaries of Indian monsoon, such as spatial distribution of rainfall, extended events of excess or deficient rainfall, relation between different climatic variables, causal drivers of Indian monsoon, and how monsoon may evolve under climate change. I am also interested in developing accurate but low-cost earth system models which will be able to simulate localized physical, climatic or hydrological processes in conjunction with global earth system models. I am also interested in development of digital twins for real-life physical systems and processes for policy evaluation. I am currently exploring how a communicable disease like covid-19 can spread in a city or a campus under various intervention strategies like total or partial lockdown and contact-tracing. Another current focus of mine is to explore remote sensing imagery using deep learning for mapping economic developments and risks, such as the detection of informal settlements that are vulnerable to natural hazards. Regarding methodology, I have been making extensive use of probabilistic graphical models, and currently exploring deep generative models. I make use of agent-based models for digital twins, and I am exploring approximate Bayesian computation for their parameter estimation. I am also exploring different notions and algorithms of causality.
Institute of Sugar Beet Research associated institute of the university of Göttingen Göttingen, Germany
Key competences: Plant pathology, Plant protection, Phenotyping, Plant resistance, Precision agriculture, Sensor and data analysis
Current research projects:
Research Group Leader
Cluster of Excellence “Machine Learning”
University of Tübingen
I have worked at the Potsdam Institute for Climate Impact Research (PIK) with Dr. Norbert Marwan on a DFG funded project (2017-2020) that focuses on the development of a new framework to work with uncertainties in climate data. Prior to that, I have worked with Prof. Dr. Bodo Bookhagen at the Institute of Geosciences, University of Potsdam, on characterising spatial patterns of hydrology in the greater Himalayas and also on estimating flow accumulation from lidar point clouds (2016-2017). I did my PhD (2015) under the supervision of Prof. Dr. Dr. h.c. Jürgen Kurths at PIK, on developing new ways to deal with uncertainties in sedimentary proxy records obtained from paleoclimate archives such as stalagmites, lake sediments, and ice cores.
I try to address relevant methodological questions in paleoclimate and climate data analysis with tools and techniques from classical statistics, machine learning, nonlinear time series analysis, and complex networks. The data are diverse and heterogeneous, arising from re-analysis, satellite radars, meteorological and hydrological station networks and paleoclimate proxy archives. Each data set has its own peculiarities and challenges, and in my experience, more often than not, we cannot apply existing methods ‘out-of-the-box’ to find the answers we wish to have. Rather, we have to adapt and extend the state-of-the-art to arrive at reliable results.
I have worked on abrupt transitions in climatic systems and how uncertainties impact our inferences on paleoclimate variability. I have also tried to unravel the interactions between climatic systems governing global temperature and how unsupervised learning can help us identify long-range interactions between oceans and rainfall. Part of my work also deals with climate networks, along with recurrence plots and extreme events.
Centre for Atmospheric Sciences IIT Delhi
Much of my research career has been in climate change. I have worked extensively in the areas of climate variability as well as climate change detection & attribution at global and regional scales. Some of my work on using climate models to understand ocean warming and interannual variability is still widely cited and used. This has led to some work on understanding regional & local drivers of climate change (especially over India) as well as in quantifying uncertainty in model projections of future climate. More recently, my work has focused on extreme events (and the mechanisms) and in the emerging field of event attribution. I am also exploring the use of conventional and AI/ML techniques in improving predictions at various time-scales – ranging from seasonal to decadal.
Department Biogeochemical Integration Max Planck Institute for Biogeochemistry, Jena
Markus Reichstein’s main research interests revolve around the response and feedback of ecosystems (vegetation and soils) to climatic variability with a Earth system perspective, considering coupled carbon, water and nutrient cycles. Of specific interest is the interplay of climate extremes with ecosystem and societal resilience. These topics are adressed via a model-data integration approach, combining data-driven machine learning with systems modelling of experimental, ground- and satellite-based observations. Markus Reichstein is Director of the Biogeochemical Integration Department at the Max-Planck-Institute for Biogeochemistry, Professor for Global Geoecology at the FSU Jena, and founding Director at the Michael-Stifel-Center Jena for Data-driven and Simulation Science. He has been serving as lead author of the IPCC special report on Climate Extremes (SREX), as member of the German Commitee Future Earth on Sustainability Research, and the Thuringian Panel on Climate. Recent awards include the Piers J. Sellers Mid-Career Award by the American Geophysical Union (2018), an ERC Synergy Grant (2019) and the Gottfried Wilhelm Leibniz Prize (2020).
Assistant Professor, Environmental Studies
Researchers from multiple disciplines at Ashoka University (biology, ecology, environmental sciences, geography and economics) use large datasets and modelling to work on different aspects of environment and sustainability. In the biological and ecological sciences, conservation geneticists use species histories, genomic footprints and habitat models to predict evolutionary trajectories under and future climatic regimes; immunologists use experimental evolution and life history analyses and physiological and molecular manipulations to assess evolutionary and ecological outcomes that have applications for understanding outcomes of population pressures on selection and on disease interactions; neuroscientists use machine learning tools to understand how neural and behavioural systems of communication may be compromised in response to ecological constraints; and ecologists use large-data modeling to determine how climate influences phenology and species interactions and predict future changes in ecosystem trajectories. In social sciences, economists use machine learning to answer questions related with labour economics, to develop crop insurance systems in light of oncoming climate change and understand discrimination based on gender and other factors. In computer science, researchers are using AI and ML to locate pollution hotspots and calibrate instruments for quantifying pollutants across space and time. Finally, researchers in the environmental studies department are creating large datasets on climate, species compositions and human activities to identify drivers for ecosystem degradation and make policy for better ecosystem management. Within environment and sustainability, Ashoka researchers are working on agriculture and crop burning systems, conservation of terrestrial and marine ecosystems and their biodiversity, labour and common property systems, and pollution drivers and solutions.
Deutsches Zentrum für Luft- und Raumfahrt (DLR), Institut für Physik der Atmosphäre,Oberpfaffenhofen
University of Bremen, Institute of Environmental Physics (IUP), Bremen
Veronika Eyring's research focuses on Earth system modeling and process-oriented model evaluation with observations, including the development and application of artificial intelligence (AI) techniques for reliable climate projections and technology assessments.
DLR Institute for Data Science
Professor Computer Vision
Faculty of Mathematics and Computer Science
Friedrich Schiller University Jena
Education and Scientific Career: