发表机构
University of California San Diego; Georgia Institute of Technology; University of California Riverside; San Diego State University（加利福尼亚大学圣迭戈分校; 佐治亚理工学院; 加利福尼亚大学河滨分校; 圣迭戈州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出叠加式潜在自编码器（SLAE），通过学习叠加共享存储，在相同内存预算下改善重构-内存权衡，降低重构误差，还提升下游分类性能，为表示压缩提供新原则。
AI 中文摘要
自编码器通常通过缩小每个潜在表示来满足严格的潜在内存预算，这会牺牲表示能力。我们提出了一个不同的问题：是否可以将多个更宽的潜在表示存储在一起？我们引入了叠加式潜在自编码器（Superposed Latent Autoencoder，SLAE），该模型在通过学习到的叠加共享存储的同时保留高容量的潜在表示。SLAE 将潜在表示转换为存储友好型代码，用随机密钥绑定这些代码，将多个代码叠加到单个内存张量中，并学习在解码前恢复每个潜在表示。在相同的存储预算下，SLAE 用可被抑制的结构化干扰替代了不可逆的维度瓶颈。在 CIFAR-10/100、SVHN、STL-10、Tiny ImageNet 以及广泛的内存预算范围内，SLAE 大幅改善了重构-内存权衡，在匹配的存储条件下，与传统自编码器相比，重构误差最多降低 56%。进一步分析表明，SLAE 的优势源于在相同存储预算下使更宽的表示变得可用。这些增益也延伸到了重构之外：在相同内存预算下，SLAE 保留的信息使下游分类性能提升最多 16.79 个百分点。我们的结果为表示压缩提出了一个新原则：不是缩小每个潜在表示，而是保留宽表示并让它们共享内存。
英文摘要
Autoencoders typically meet tight latent-memory budgets by making each latent representation smaller, sacrificing representational capacity. We ask a different question: can multiple wider latents be stored together instead? We introduce the Superposed Latent Autoencoder (SLAE), which preserves high-capacity latent representations while sharing storage through learned superposition. SLAE transforms latents into storage-friendly codes, binds them with randomized keys, superposes multiple codes into a single memory tensor, and learns to recover each latent before decoding. Under the same storage budget, SLAE replaces irreversible dimensional bottlenecks with structured interference that can be suppressed. Across CIFAR-10/100, SVHN, STL-10, Tiny ImageNet, and a wide range of memory budgets, SLAE substantially improves the reconstruction--memory tradeoff, reducing reconstruction error by up to 56% over conventional autoencoders at matched storage. Further analysis shows that SLAE's advantage comes from making wider representations usable under the same storage budget. These gains also extend beyond reconstruction: the information preserved by SLAE improves downstream classification by up to 16.79 percentage points under the same memory budget. Our results suggest a new principle for representation compression: instead of making every latent smaller, keep representations wide and let them share memory.
