关联特征条件化的集合-时序Transformer用于跨会话皮层内运动解码
Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
浏览论文内容
中文总结 AI 辅助
提出APST，一种冻结权重的集合-时序Transformer，利用闭式关联特征适应新会话，在跨会话皮层内运动解码中显著提升速度解码精度。
中文摘要 AI 辅助
皮层内运动解码器在跨会话时性能会下降，因为记录到的神经元集合会发生变化，且持续存在的神经元其放电与行为之间的关系也可能改变。现有方法大多在每个新会话上更新网络权重，或依赖未标注的活动数据，但这并不能直接揭示上述变化。我们提出了APST，一种关联特征条件化的集合-时序Transformer（Association Profile-conditioned Set-Temporal transformer），它在所有网络权重冻结的情况下适应新会话。通过少量有标注的校准试验，APST以闭式解计算每个神经元放电与行为关系的四维关联特征。这些特征对集合注意力编码器进行条件化，该编码器可接受任意数量和顺序的神经元，随后接一个因果Transformer用于流式解码。在来自两只猴子的留出DANDI688会话上，APST达到了速度$R^2$为$0.78$和$0.81$，而仅使用神经活动的变体为$0.40$和$0.58$，并且APST匹配或超过了在同一试验上微调的RNN。在FALCON私有留出评估中，它在M1、M2和H1上分别达到了$R^2$为$0.65$、$0.42$和$0.44$。
英文摘要
Intracortical motor decoders degrade across sessions because the set of recorded units changes and persisting units can alter how their firing relates to behavior. Most existing methods update network weights on each new session or rely on unlabeled activity, which does not directly reveal such changes. We present APST, an Association Profile-conditioned Set-Temporal transformer that adapts to new sessions with all network weights frozen. From a few labeled calibration trials, APST summarizes how each unit's firing relates to behavior in a four-dimensional association profile computed in closed form. The profiles condition a set-attention encoder that accepts any number and order of units, followed by a causal transformer for streaming decoding. On held-out DANDI688 sessions from two monkeys, APST reaches velocity $R^2$ of $0.78$ and $0.81$, versus $0.40$ and $0.58$ for a variant that uses neural activity alone, and matches or exceeds an RNN fine-tuned on the same trials. On FALCON private held-out evaluation, it attains $R^2$ of $0.65$, $0.42$, and $0.44$ on M1, M2, and H1.
发表机构
- The University of Hong Kong（香港大学）
- Southern University of Science and Technology（南方科技大学）
机构由 AI 辅助整理，请以论文原文为准。
