Research map / Learning theory and algorithms / Clustering
Fast and accurate spike sorting of high-channel count probes with KiloSort
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Low-rank (SVD, rank 3) template representation with matrix-product-then-convolution filtering | KiloSort template matching · Runtime analysis | computational operations: reduces operations by factor Nchan/Nrank | Helped |
| Batch-based online template optimization on GPUs instead of KlustaKwik-style optimization | KiloSort · Hybrid ground truth, 32 and 128 channels | runtime (minutes): 10 and 29 min vs ~480 and 10-20 thousand min for KlustaKwik | Helped |
| Full KiloSort vs KlustaKwik | KiloSort · Hybrid ground truth | fraction of units with score > 0.9: 53% vs 35% | Helped |
| Oracle greedy merges applied after sorting | KiloSort · Hybrid ground truth | fraction of units with score > 0.9: 53% to 69% | Helped |
| Automated post-hoc template merging based on continuity of 2D feature-space projections | KiloSort · Hybrid ground truth | manual merge operations needed: reduces manual operations substantially (no numeric value given) | Helped |
| Operating on raw data with overlapping-spike template matching instead of PCA compression | KiloSort · Hybrid ground truth | misses (miss rate): qualitative: relative improvement driven by fewer misses | Helped |
| Low-rank spatiotemporal SVD (3 components) waveform model instead of common temporal PCs per channel | Mean spike waveform reconstruction · One dataset (Fig. 2b) | residual waveform variance: ~5 times less residual variance | Helped |
Closest papers by meaning
- SimSort: A Data-Driven Framework for Spike Sorting by Large-Scale Electrophysiology Simulation (NeurIPS 2025 · similarity 0.90)
- ActSort: An active-learning accelerated cell sorting algorithm for large-scale calcium imaging datasets (NeurIPS 2024 · similarity 0.83)
- SpikeSSL: A Universal Spike Inference Framework with Dynamics-Informed State-Space Layers (NeurIPS 2026 · similarity 0.82)
- S$^2$M-Former: Spiking Symmetric Mixing Branchformer for Brain Auditory Attention Detection (NeurIPS 2025 · similarity 0.82)
- convSeq: Fast and Scalable Method for Detecting Patterns in Spike Data (ICML 2024 · similarity 0.82)
- Spike Distance Function as a Learning Objective for Spike Prediction (ICML 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.