Research fellow positions - 2022

Research agenda

Projects and positions

At this moment we have open research positions (both PhD and postdoc) to work on AI, machine learning, model inversion and causality in the following projects:

1. Microsoft Research (MSR): "Causal4Africa: Causal inference to understand food security” [1 postdoc *or* 1 PhD]

Causal inference eXplainable AI Climate/social sciences
Application form: here
Enquiries: Prof. Gustau Camps-Valls
The Causal4Africa project will investigate the problem of food security in Africa from a novel causal inference standpoint. The problem is of pressing urgency given the severe situation in the continent and in the Horn of Africa in particular. Food security is at an unprecedented risk level triggered by continuous drought events, complicated interactions between food prices, energy inflation, and lack of (or timid) humanitarian aid/action, along with disrupting conflicts and migration flows. Complex, multivariate, multiscale, nonlinear cause-effect relations and legacy effects happen in the wild, which are difficult to disentangle and understand by canonical data science approaches. The problem needs advanced causal discovery and causal evaluation algorithms to discover the causal relations and to evaluate the likelihood and potential consequences of specific interventions. We will collect a large variety of observational data: human displacement, food prices and availability, energy, conflicts, humanitarian operations as well as environmental and climatic data over the last decade. Causal discovery approaches and cause-effect impact assessment methods will help in understanding the coupled system and the impact of humanitarian interventions on food security levels from the IPC classification. We will employ storylines, causal discovery methodologies [ISP toolboxes, Runge et al 2019,] and methods in EconML (e.g. doubleML) to estimate heterogeneous treatment effects from observational data. We recall the dictum “Models without data are fantasy but data without models are chaos” and will rely on (and interact with) expert knowledge to guide and evaluate our algorithmic approaches and solutions. We anticipate that our causal algorithmic approaches, illustrated in this project for food security analysis, may help adaptation to climate change. The methodological developments may also find application in other related but different domains.
This is a joint collaborative project between the University of Reading (led by Ted Shepherd), Universitat de València (led by Prof. Camps-Valls, ISP), and Microsoft Research (led by Alberto Arribas, head of the Europe Lead for Corporate Environmental Sustainability Science). The successful candidate will be based in the Universitat de València, but will actively collaborate and visit the other groups and MSR (Cambridge, UK).
Who should apply? only if you are knowledgeable in causal inference, and interested in climate and social sciences

2. PROMETEO AI4CS: "AI for Complex Systems" [2 postdocs, 3 PhD students]

Deep learning Model inversion eXplainable AI Causality
Application form: here
Enquiries: Prof. Gustau Camps-Valls or Dr. Maria Piles
Our vision in AI4CS is to develop novel artificial intelligence methods to model and understand complex systems, and more specifically the visual brain, Earth and climate systems and the biosphere-anthroposphere interactions. A perfect storm is over us: (i) an ever increasing amount of observational and sensory data, (ii) improved high resolution yet mechanistic models are available, and (iii) advanced machine learning techniques able to extract patterns and identify drivers from data. In the last decade, machine learning models have helped to monitor, predict and forecast all kind of variables and parameters of interest from observational data. They help in quantifying visual stimuli, to monitor land, oceans, and the atmosphere, as well as to study socio-economic variables at different scales and spheres. Current approaches, however, face three important challenges: (1) they cannot deal efficiently with the particular characteristics of data, (2) they do not respect the most elementary laws of physics, and (3) they just interpolate but nothing fundamental is learned from data.
In AI4CS we tackle these three problems by designing algorithms able to deal with huge amounts of complex, heterogeneous, multisource, and structured data. Firstly, a new generation of targeted AI methods to improve efficiency, prediction accuracy, and uncertainty quantification and error propagation. Secondly, we push the boundaries of a new family of hybrid physics-aware machine learning models that encode physical knowledge about the problem, constraints, inductive biases and domain knowledge, with the goal of attaining self-explanatory models learned from empirical data. Finally, the project deals with learning graphical causal models to explain the potentially complex interactions between key observed variables, and discover hidden essential drivers and confounding factors. The AI4CS project vision thus seizes the fundamental problem of moving from correlation to dependence and then to causation through data analysis. The theoretical developments are guided by the inherent ventures of modeling and understanding complex systems at different spatio-temporal resolutions, spheres and interactions.
The positions:
  1. PhD1: Interplay between human visual system and deep learning
  2. PhD2: Hybrid machine learning modeling: improved parameter retrieval and data-driven physics discovery
  3. PhD3: Causal inference & Extreme event detection and impact attribution in climate sciences
  4. Postdoc 1: Modes of variability and causal discovery, dimensionality reduction in spatio-temporal data
  5. Postdoc 2: Causal inference & Extreme event detection and impact attribution in climate sciences
The researchers teamed up in AI4CS are led by the Image and Signal Processing (ISP) group at the Universitat de València, and several outstanding researchers join from different institutions to collaborate actively together. The AI4CS research team is formed by 13 senior researchers and very active. The successful candidates will be based in the Universitat de València, Spain, and will be supervised by both members of the ISP and from other national and international institutions in the consortium. This will imply frequent teleconfs, possibility of teleworking, and research stays abroad. This is an opportunity to hone team skills while collaborating with a team of researchers.
Who should apply? only if you are knowledgeable in machine learning, deep learning & causal inference, and interested in Earth, climate and social sciences

Your Profile

  1. Experience in machine learning, deep learning, image processing, statistics, Bayesian inference
  2. We love interdisciplinarity! Interested in remote sensing, Earth sciences, climate science
  3. Experienced in scientific interpretation and analysis of data
  4. Experienced with or convincing motivation to enter a leadership position
  5. Familiar with modelling, model-data-fusion and machine learning
  6. Excellent quantitative skills (e.g. data analysis, modelling)
  7. Strong programming skills in Python/Julia/R
  8. Proven record of scientific publications in international peer-reviewed journals
  9. PhD in maths, physics, or computer science
  10. Critical and organized sense for data analysis
  11. Maturity and commitment
  12. Strong communication, presentation and writing skills are a big plus
  13. Collaborative team player

Your tasks

  1. Lead research along the project specific topics
  2. Collaborate with an international team of researchers
  3. Postdocs should (co)advise PhD students
  4. Publish in international peer-reviewed journals and conference venues

Why ISP? Context and offer

The successful candidates will be based in the Universitat de València, Spain. Prof. Camps-Valls is the coordinator of the Image and Signal Processing (ISP) group. The group is devoted to the development of machine learning and signal processing techniques for remote sensing image processing, Earth observation data analysis and the Geosciences. Several topics are treated in our research group and projects: regression, causality & information theory, Earth observation data analysis, physics-aware machine learning, generative modeling, eXplainable AI and feature ranking, and anomaly detection. The ISP group is the place to be:
  1. Truly interdisciplinary projects: machine learning + climate/geo sciences + neuroscience + socio-economics
  2. Work in cutting-edge machine learning to tackle relevant, challenging societal, environmental and climate problems
  3. Postdocs may supervise outstanding students and lecture in top EU master is possible, yet not mandatory.
  4. Access to high performance computing facilities and clusters
  5. Part-time job and teleworking are generally acceptable
  6. Flexibility to work on side projects with companies and international organizations (see list of collaborators)
  7. We care about diversity and the gender issue!
  8. Active group, engaged in vibrant AI networks like ELLIS, i-AIDA and ESA PhiLab and with cool collaborators
  9. Very friendly, interactive and international working environment
  10. Salary according to UV scales + health insurance + travel money. Excellent cost-of-living index = 55