Climate forecasting and geospatial ML at IBM Research
Applied ML research at IBM Research: climate forecasting models deployed into IBM PAIRS Geoscope, published work, and a provincial COVID-19 risk dashboard.
Search
Writing, projects, publications, talks, and courses.
30 results for “Research”
Session 6.7: How to get AI to teach you writing & research
Co-hosting with Prof Phakeng on a different way to think about AI: using it to strengthen your own thinking rather than replace it. We explore how AI can challenge your assumptions, improve your writing, and deepen your understanding of a research problem.
Apache Flink
Work on apache flink across projects, research and writing.
Apache Kafka
Work on apache kafka across projects, research and writing.
Artificial intelligence
Work on artificial intelligence across projects, research and writing.
Bayesian filtering
Work on bayesian filtering across projects, research and writing.
Capacity building
Work on capacity building across projects, research and writing.
Climate risk and geospatial machine learning
Climate risk and geospatial ML: seasonal forecasting research, models in a petabyte scale platform, and flood forecasting for underwriting.
Computer vision
Computer vision research: learning level set image segmentation with echo state networks and recurrent models, plus an object following robot.
Dask
Work on dask across projects, research and writing.
Data science
Applied data science in banking, telecoms, research and higher education, and writing on building the capability around the models.
Databricks
Work on databricks across projects, research and writing.
Decision support systems
Work on decision support systems across projects, research and writing.
Deep learning
Work on deep learning across projects, research and writing.
Django
Work on django across projects, research and writing.
Docker
Work on docker across projects, research and writing.
Echo state networks
Work on echo state networks across projects, research and writing.
Echo state networks for level set image segmentation
MSc research at Wits: can recurrent models learn the iterative evolution of a level set segmentation? Echo state networks against trained RNN, LSTM and GRU.
Ensemble Kalman filter
Work on ensemble kalman filter across projects, research and writing.
Extract, transform, load
Work on extract, transform, load across projects, research and writing.
Geospatial machine learning
Work on geospatial machine learning across projects, research and writing.
Image segmentation
Work on image segmentation across projects, research and writing.
Insurance
Work on insurance across projects, research and writing.
Kalman filtering
Work on kalman filtering across projects, research and writing.
Kubernetes
Work on kubernetes across projects, research and writing.
Level set method
Work on level set method across projects, research and writing.
Machine learning
Work on machine learning across projects, research and writing.
Machine learning for seasonal climate forecasting
IBM Research work on forecasting 2m temperature and precipitation weeks to months ahead with machine learning, published at NeurIPS 2020 CCAI and EGU 2022.
Mathematical optimisation
Work on mathematical optimisation across projects, research and writing.
ML-based Probabilistic Prediction of 2m Temperature and Total Precipitation
Proposes a daily probabilistic model that forecasts 2 m temperature and total precipitation globally, aimed at the skill gap between weather and seasonal forecasting. Physics based ensembles, climate modes and recent climatology are combined as inputs to Extreme Gradient Boosting, U-Net and Natural Gradient Boosting models. The authors report that it consistently outperforms both ECMWF 46 day forecasts and climatology. Co-authored across IBM Research in South Africa, Kenya, Brazil and the US.