Python development for encoding/decoding methods in SNNs

Context

Spiking Neural Networks (SNNs) represent a biologically inspired form of neural computation that processes information using discrete spikes over time. However, most available datasets are based on continuous (floating point) data, which makes them incompatible with SNNs without suitable preprocessing. To bridge this gap, we are developing a dedicated repository that implements standard encoding schemes (e.g., rate coding, temporal coding) to convert continuous data into spike trains, as well as decoding methods to interpret SNN outputs.

Tasks

In this position, you will design and implement a Python-based repository of commonly used encoding and decoding strategies for spike signals. This will include integrating existing methods from the literature as well as ensuring the codebase is modular, extensible, and well-documented. The resulting tool will support future projects using SNNs and enable easier benchmarking and deployment on neuromorphic systems.Your tasks will include:

• Research and implement standard spike encoding and decoding techniques
• Design a modular and user-friendly Python repository
• Validate the methods on representative datasets

Requirements

• Solid programming skills in Python
• Understanding of Artificial Neural Networks
• (Optional) Experience with Spiking Neural Networks, neuromorphic computing, or related topics