无困惑参数t-SNE
Perplexity-free Parametric t-SNE
摘要:
t-分布随机邻域嵌入(t-SNE)算法是一种广泛使用的降维(DR)方法。其非参数性质和令人印象深刻的功效促使其参数扩展。然而,它是有界的用户定义的困惑参数,限制其DR质量相比,最近开发的多尺度的困惑自由的方法。因此,本文提出了一种多尺度参数t-SNE方案,解除了困惑调整和深层神经网络实现映射。它产生可靠的嵌入与样本外的扩展,竞争力的最佳困惑调整方面的邻域保护多个数据集。
英文摘要:
The t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm is a ubiquitously employed dimensionality reduction (DR) method. Its non-parametric nature and impressive efficacy motivated its parametric extension. It is however bounded to a user-defined perplexity parameter, restricting its DR quality compared to recently developed multi-scale perplexity-free approaches. This paper hence proposes a multi-scale parametric t-SNE scheme, relieved from the perplexity tuning and with a deep neural network implementing the mapping. It produces reliable embeddings with out-of-sample extensions, competitive with the best perplexity adjustments in terms of neighborhood preservation on multiple data sets.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。