Skip to content
View Polichinel's full-sized avatar

Block or report Polichinel

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Polichinel/README.md

πŸ’« About Me:

πŸ’Ό Senior Researcher At the Peace Research Institute Oslo (PRIO), where I lead advanced research initiatives focused on conflict forecasting.

πŸš€ Team Leader Team Leader for Model Development and Deployment (MD&D) at the Violence Impact and Early Warning System (VIEWS), delivering actionable insights to humanitarian organizations.

πŸ” Current Focus Areas:

  • HydraNet: Pioneering tempo-spatial learning using U-net + LSTM architectures to enhance conflict forecasting.

  • VIEWS Forecasting Pipeline: Developing the next-generation pipeline for more flexible, robust, and transparent conflict forecasts

  • VIEWS-FAO Conflict Return Periods: Pilot project of estimating the recurrence of highly destructive conflict events in collaboration with FAO to inform parametric insurance.

  • Actor-Level Conflict Escalation Patterns: Applying cutting-edge LLMs and newswire text analysis to decode conflict escalation dynamics at the actor level.

🎯 Driving Force: Trying really hard to generate actionable insights that enables early action, aiming to reduce the human toll of violent conflicts.

πŸ’‘ Problem-Solving Philosophy: Fuel up on energy drinks, code, code, code, dive deep, get confused, embrace the confusion, get confused, repeat

πŸ’¬ Ask Me About:

  • Forecasting violent conflict with Machine Learning.
  • Tempo-spatial modeling with neural networks.
  • The role (current and future) of AI and machine learning in peace and conflict processes.
  • Transparency, interpretability, and causality in AI-driven early warning systems for peace and conflict.
  • Collaborations or partnerships in conflict prediction.

🌐 Socials:

LinkedIn

πŸ’» Tech Stack:

LaTeX Markdown Python R Google Cloud Anaconda Prefect Gimp Inkscape Keras Matplotlib NumPy Pandas PyTorch scikit-learn Scipy TensorFlow Plotly GitHub Actions Git GitHub

πŸ“Š GitHub Stats:



Pinned Loading

  1. from_headlines_to_hotspots from_headlines_to_hotspots Public

    Jupyter Notebook 3 1

  2. HydraNet_001 HydraNet_001 Public

    Repsetory for ConflictNet. A conflict forecasting network based on a recurrent - approximate baysian - Unet.

    Jupyter Notebook 1

  3. the_currents_of_conflict the_currents_of_conflict Public

    Updated version

    Jupyter Notebook

  4. prio-data/viewsforecasting prio-data/viewsforecasting Public

    Jupyter notebooks and python scripts for performing the ViEWS monthly forecasts

    Python 13 1

  5. prio-data/views_pipeline prio-data/views_pipeline Public

    VIEWS forecasting pipeline for monthly prediction runs. Includes MLops and QA for all models/ensembles.

    Jupyter Notebook 3 3

  6. prio-data/VIEWS_FAO_index prio-data/VIEWS_FAO_index Public

    Repo for the VIEWS FOA conflict index project

    Jupyter Notebook