quantum_hackathon_2024

Champion of Quantum Hackathon organized by QWorld and School of Code and sponsored by SandboxAQ to build apps to tackle the UN’s 17 Sustainable Development Goals!

Current Problem:

Despite collecting terabytes of environmental and ecological data around the globData complexity often overwhelm Conservation experts, decision-makers and the general public , leading to delayed responses and apathy, especially in critical ecosystems where every moment can count

Our Solution:

An easy-to-use app that bridges the gap between complex data and actionable insights making future impact visible in hope to prevent ecological collapse.

It:

• Leverages quantum technology to enhance data processing and analysis capabilities.
• Provide actionable intelligence to support biodiversity conservation efforts.
• Identify and analyse key factors affecting biodiversity, such as climate change.
• Create predictive models to forecast the impacts of climate change on ecosystems.
• Facilitate data collection and sharing among conservationists and researchers, and eventually citizen scientists.

Why Quantum Theory?

Classical Technique Limitations:

• Complex ecological interactions
• Multiple environmental variables
• Non-linear relationships
• Computational intensity
• Data projection accuracy

Quantum Advantage

• Better handling of uncertainty
• Potentially greater accuracy in long-term predictions
• Inclusion of more environmental variables
• Faster processing of large datasets

Penny Lane

A framework that connects quantum circuits and algorithms with classical machine learning algorithms A bridge between classical ML and quantum computing

Quantum Variational Circuit Approach

• Implemented a Variational Quantum Circuit (VQC) for biodiversity prediction
• Used 4-qubit system with 2 parameterised layers
• Combined quantum encoding with classical optimisation
• Leverage quantum superposition for complex pattern recognition in multi-dimensional data

Overall approach

Figma - ideation, planning

• started with UN17 goals and chose a topic based on something we were both passionate about - sustainability

Python - rich ecosystem of libraries, rapid development

Datasets

• combined BioTIME - global biodiversity measurements
• cached data from meteostat api - historical weather data from stations around the world

Steps:

• Process and format the data
• Train a predictive model and develop the implementation of the chosen algorithm
• Refine/rebuild
• measure and evaluate the performance

You could view our project through here.

https://www.canva.com/design/DAGWsavE690/zAsq1nuy68_N--RfWjCykw/view?utm_content=DAGWsavE690&utm_campaign=designshare&utm_medium=link&utm_source=editor