Research map / Label-efficient and robust learning / Self-supervised and contrastive
TreeMoCo: Contrastive Neuron Morphology Representation Learning
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 |
|---|---|---|---|
| add batch normalization in neuron encoder (ablation) | TreeMoCo · BIL-6 KNN | KNN accuracy (%): 69.14 -> 75.00 | Helped |
| add batch normalization in projection head (ablation) | TreeMoCo · BIL-6 KNN | KNN accuracy (%): 75.00 -> 76.56 | Helped |
| add MLP to neuron encoder (supervised) (ablation) | Tree-LSTM encoder · BIL-6 | accuracy (%): 80.86 -> 83.98 | Helped |
| add MLP to neuron encoder (supervised) (ablation) | Tree-LSTM encoder · JML-4 | accuracy (%): 54.29 -> 57.14 | Helped |
| add MLP to neuron encoder (supervised) (ablation) | Tree-LSTM encoder · ACT | accuracy (%): 43.16 -> 50.53 | Helped |
| add batch normalization to encoder (supervised) (ablation) | Tree-LSTM encoder with MLP · ACT | accuracy (%): 50.53 -> 63.16 | Helped |
| use 2 LSTM cells instead of 1 (ablation) | Tree-LSTM encoder with BN and MLP · BIL-6 | accuracy (%): 84.77 -> 85.55 | Helped |
| use 2 LSTM cells instead of 1 (ablation) | Tree-LSTM encoder with BN and MLP · JML-4 | accuracy (%): 58.57 -> 64.29 | Helped |
| use 2 LSTM cells instead of 1 (ablation) | Tree-LSTM encoder with BN and MLP · ACT | accuracy (%): 63.16 -> 62.11 | Hurt |
| fine-tune pre-trained encoder with labels (ablation) | TreeMoCo · BIL-6 | accuracy (%): 78.91 -> 83.98 | Helped |
| fine-tune pre-trained encoder with labels (ablation) | TreeMoCo · ACT | accuracy (%): 55.79 -> 68.42 | Helped |
| unsupervised TreeMoCo instead of supervised TRNN (ablation) | TreeMoCo frozen KNN · BIL-6 | accuracy (%): 76.56 -> 78.91 | Helped |
Closest papers by meaning
- MorphGrower: A Synchronized Layer-by-layer Growing Approach for Plausible Neuronal Morphology Generation (ICML 2024 · similarity 0.87)
- Neuron Platonic Intrinsic Representation From Dynamics Using Contrastive Learning (ICLR 2025 · similarity 0.85)
- Reproducibility of predictive networks for mouse visual cortex (NeurIPS 2024 · similarity 0.84)
- Joint Learning Neuronal Skeleton and Brain Circuit Topology with Permutation Invariant Encoders for Neuron Classification (AAAI 2024 · similarity 0.84)
- Learning to cluster neuronal function (NeurIPS 2025 · similarity 0.83)
- Learning Multimodal Volumetric Features for Large-Scale Neuron Tracing (AAAI 2024 · similarity 0.83)
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.