Tag
Handwired 是一个趣味项目，可将普通 JavaScript 代码转换为神经网络，无需训练即可获得与原程序完全一致的输出，并且完全可微，全部在浏览器中运行，无需 GPU。
This paper introduces DFSC, a PyTorch environment built around a Mittag-Leffler Spectral Layer (MLSL) that enables error-controlled, differentiable fractional propagation in hybrid neural models. It separates known fractional dynamics from data-driven corrections, supports batched/dense/sparse operator paths, and provides certified error bounds in certain regimes.
This blog post explores a technique to make legacy Fortran simulation code differentiable using LFortran, Enzyme, and Tesseract, allowing automatic differentiation and integration with JAX for use in machine learning pipelines.
Introduces Automatically Differentiable Nonlinear Tensor Networks (ADNTNs) for compressing deep neural network layers via small core tensors, achieving high compression ratios while maintaining accuracy.
This article explores neural cellular automata as a computational model inspired by biological morphogenesis and regeneration, demonstrating how simple local rules can lead to complex global behaviors.
This paper presents 'Additive Atomic Forests,' a framework for simultaneous symbolic recovery of functions and their antiderivatives using derivative algebra and self-expanding atom libraries. The method achieves strong performance on classification benchmarks and Feynman symbolic regression tasks while offering interpretable results.
This paper presents Block-Wise Differentiable Sinkhorn Attention, a method for efficient long-context balanced entropic optimal transport attention on TPU hardware. It introduces a tail-refinement surrogate for exact differentiation, proving an efficient backward pass schedule and demonstrating significant improvements in Pfam sequence alignment reconstruction.