Jens E. Pedersen

See how leaky integrators provide scale-space covariance for SNNs, boosting event-based tracking by 42% when initialized with spatio-temporal priors.

Submit your open-source neuromorphic projects for transparent community review and recognition through the ONR Program.

Implementing ROSBag format support in Faery demonstrates how compiling Rust modules into a Python API ensures high-performance event data decoding.

The Tonic library standardizes event-based data loading and transformation, providing a PyTorch-compatible pipeline that accelerates SNN model training.

Comparing neuromorphic and conventional ML workflows reveals a massive infrastructure gap—and big opportunity to accelerate the field.

Following a landmark panel, leading projects like BindsNET, Brian, GeNN, and snnTorch unite to form the inaugural Open Neuromorphic Collaboration Network, putting our strategic vision into action.

Nine neuromorphic tool maintainers navigate open-source funding, standardize edge deployments, and successfully balance commercial scale with biological realism.

Why 'open' matters and where we want to take the Open Neuromorphic community

Learn how extracting computational graphs with torch.fx allows the Neuromorphic Intermediate Representation (NIR) to bridge PyTorch models and SNN hardware.

Discover the architectural decisions behind Faery v0.3.0, including its new MP4 conversion pipeline, dynamic CLI reflection, and Rust-based build challenges.

See how the Faery API handles event-stream regularization and custom CLI parsers to efficiently render uncalibrated neuromorphic event data into MP4 video.

The Faery library's CLI design uses UNIX-style piping to simplify event camera data conversions and complex, real-time filtering pipelines.

Discover how optimizing recurrent SNN loops with JAX's scan operation yields a 5x speedup over unrolled functions without needing low-level Pallas code.

Faery's stream-based API and the Neuromorphic Intermediate Representation (NIR) enable deploying models to Innatera's mixed-signal edge chips.

Explore how to use GitHub Actions and cibuildwheel to automate cross-platform compilation and deployment of the Rust-backed Faery event processing library.

See how the Neuromorphic Intermediate Representation (NIR) enables seamless SNN model deployment across Norse, SynSense Speck, and SpiNNaker2 hardware.

A hybrid Spiking Neural Network and event-based Expectation Maximization pipeline deployed on the BrainChip Akida tracks pedestrians with a ~6W power footprint.

Discover the fastest Spiking Neural Network (SNN) frameworks for deep learning-based optimization. Performance, flexibility, and more analyzed in-depth