The Environmental Intelligence for Global Change Lab at Politecnico di Milano invites applications for 3 PhD scholarships (3 years each) and 3 postdoctoral positions (24 months each) distributed over three research topics. All positions are funded by the Fondo Italiano per la Scienza (the Italian version of the European ERC grants) on the research project BERLIN – BEhavioRaL INtelligence for evidence-based adaptation policies, which is expected to start in Fall 2025.
BERLIN aims to improve the understanding and modeling of how the behavior of human agents influences the complex dynamics of coupled human-natural systems. To achieve this objective, BERLIN will construct data-driven behavioral models of the agents’ intentions and preferences by leveraging the recent advances in Machine Learning, which allow exploiting the full potential of the unprecedented availability of big observational data. Moreover, BERLIN will address risk assessment by delving into stakeholders’ experiences and preferences from a triple-loop approach (risk awareness, risk perception, and risk adaptation) to explore how Social Learning and user-driven indicators can reinforce the model-based exploration of adaptation policies. Climate narratives and storylines will be constructed to represent self-consistent past events and the plausibility of different adaptation pathways. The integration of these transdisciplinary research efforts will support the development of a behaviorally explicit global hydrologic model for supporting rigorous retrospective assessments of the observed agents’ behaviors as well as the development of reliable and credible projections about the future coevolution of coupled human-natural systems. To provide an evidence-based foundation for the identification of adaptation policies at the local scale, BERLIN will consider different Climate Change Hotspots including semiarid regions, river deltas, and snow-dependent river basins. As a result, BERLIN will open a new path for modeling human behaviors, supporting the codesign of local adaptation strategies as well as the achievement of regional-to-global scale targets related to water management, energy and food security as part of the SDGs.
TOPIC 1 – Inverse Reinforcement Learning for modeling multi-purpose reservoir operations
This research topic will focus on developing Inverse Reinforcement Learning algorithms for the automatic extraction of intentions (i.e., main operating objectives representing the different stakeholders’ demands) and preferences (i.e., tradeoff balancing the confronted demands) associated with the observations of water reservoir levels and releases.
Qualifications for these positions include a Ph.D./MSc in computer and/or automation engineering, artificial intelligence, applied mathematics or related fields. Strong numerical and computational skills are required as well as English language skills both in oral and written communication. Strong programming skills in Python or MATLAB are required.
TOPIC 2 – Modeling climate risk perception via Social Learning
This research topic will focus on understanding how stakeholders perceive and respond to changing environmental and climatic conditions, form expectations about the uncertain future, and adapt their behavior under different scenarios. For this purpose, stakeholders’ mapping will be conducted in tandem with semi-structured interviews, group discussions, and collective workshops to identify climate risk narratives and behaviors, followed by a structured questionnaire for evaluating roles, functions, influence, and interactions between stakeholders when addressing climate risk.
Qualifications for these positions include a Ph.D./MSc in social sciences with a robust qualitative research and stakeholders analysis background, including environmental sociology, human geography, political ecology and related fields. Alternatively, candidates experienced in water governance, hydrosocial research and behaviour modelling are also encouraged to apply. Inter- and transdisciplinarity expertise together with strong social computing, system dynamics and mixed methods are required as well as advanced English language skills both in oral and written communication.
TOPIC 3 – Coupled Human-Natural systems coevolution at regional-to-global scales
This research topic will focus on investigating local adaptation policies in different Climate Change Hotspots and their potential upscaling to achieve regional-to-global scale targets. The selected hotspots include semiarid regions (i.e. Southern Africa and Spain), river deltas (i.e. Vietnam), and snow-dependent river basins (i.e. Italian Alps and California) that are prone to severe impacts from projected climate conditions, and display a wide range of socio-economic development and adaptive challenges.
Qualifications for these positions include a Ph.D./MSc in water resources engineering, hydrology, or a related field of environmental engineering. Alternatively, candidates having a background in machine learning, applied mathematics, and automation and control are also encouraged to apply. The successful candidate is experienced with water resources systems operation design, multiobjective optimization, real time control of water resources systems. Strong numerical and computational skills are required as well as English language skills both in oral and written communication.
The application package should be sent using the following forms:
Interviews will begin on 15 May 2025.
For informal enquiries, please contact Prof. Matteo Giuliani.