arXiv:2606. 30497v1 Announce Type: cross Abstract: We present a comparative study of CUDA optimization strategies applied to forward and backward propagation in a shallow neural network.
arXiv:2608. 08961v1 Announce Type: new Abstract: AI training's rising resource intensity is straining electricity supplies and carbon budgets, motivating systematic study of memory-efficient training on constrained hardware.
arXiv:2606. 06539v1 Announce Type: cross Abstract: Forward-Forward (FF) learning [Hinton, 2022] replaces backpropagation with strictly layer-local goodness updates.
arXiv:2607. 01984v1 Announce Type: cross Abstract: Newer lightweight convolutional neural networks are often presented as improving predictive performance and deployment efficiency, but such claims require controlled evaluation.
arXiv:2606. 17249v1 Announce Type: cross Abstract: The dominant trajectory of modern machine learning has been to scale up: larger models, larger accelerators, larger memory budgets.
The paper introduces PotentialEP, a backpropagation‑free method for training generative models at CIFAR scale using Equilibrium Propagation. It shows that a predictive coding network with a small linear injection can relax to a state that inherently carries the gradient needed for sampling, allowing an Energy Matching model trained with backpropagation to be sampled by relaxation alone with negligible loss in FID. Moreover, the authors demonstrate that training end‑to‑end with Equilibrium Propagation yields FID scores comparable to backpropagation, marking the first competitive CIFAR‑10 generative model trained without backpropagation.
arXiv:2606. 20005v1 Announce Type: cross Abstract: Attention distillation, which trains one attention distribution to match another by minimizing their Kullback-Leibler (KL) divergence, is widely used in knowledge distillation, model compression, continual learning, and sparse-attention LLM training.
arXiv:2607. 14568v1 Announce Type: cross Abstract: A companion study ran a 35B mixture-of-experts model on a 2011 NVIDIA Tesla C2075 (Fermi, sm_20, 6GB) as a GPU-prefill/CPU-decode hybrid, because the 4-bit model did not fit in device memory (arXiv:2606.
arXiv:2506. 21833v2 Announce Type: replace Abstract: Forward-mode automatic differentiation (FmAD) and zero-order (ZO) optimization are increasingly proposed as memory-efficient, backpropagation-free alternatives for large language model (LLM) fine-tuning, yet their benefits are typically evaluated only against standard backpropagation (BP), omitting memory-efficient variants such as activation checkpointing.
AI training's rising resource intensity is straining electricity supplies and carbon budgets, motivating systematic study of memory-efficient training on constrained hardware. We benchmark five gradient optimizers (SGD, Adam, Adagrad, Adadelta, and Conjugate Gradient Descent) under three memory strategies (standard training, gradient checkpointing, and gradient accumulation) across four transformer architectures (ViT, ModernBERT, Llama 3.
FlashBoB introduces an I/O‑efficient algorithm for exact backward‑over‑backward (BoB) in softmax attention, enabling precise second‑order differentiation without large intermediate tensors. By exploiting a hierarchical affine structure, the method confines computation to on‑chip tiles and limits off‑chip memory traffic, achieving θ(N² d²/M) HBM usage. Experiments show FlashBoB scales to sequence lengths of 262K on a single A100 GPU, outperforming prior exact baselines and FlashBack by up to 6.3×.
The paper introduces an implicit-perturbation zeroth-order (IPZO) architecture for fine-tuning spiking transformers on in‑memory computing (IMC) accelerators. By generating perturbations only for spike‑activated weight rows and combining them with IMC weighted sums, the design eliminates costly read‑modify‑write operations and reduces the hardware footprint of random number generators. An address‑driven XOR recombination scheme (PGU‑XOR) further mitigates spatial correlations, achieving near‑software accuracy while cutting perturbation energy by up to 50% compared to conventional explicit perturbation methods.