A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Benjamin Scellier, Yoshua Bengio
Equilibrium Propagation is a learning framework for energy-based models that computes gradients like backpropagation using only local perturbations, making brain-like implementation more plausible.
Unknown
This paper introduces Lora-fa, a memory-efficient low-rank adaptation method that reduces the memory footprint of fine-tuning large language models by avoiding the need to store full-rank gradients.
Jarrod R. McClean, Sergio Boixo, Vadim N. Smelyanskiy, et al.
Random parameterized quantum circuits cause exponentially vanishing gradients (barren plateaus) in hybrid quantum-classical training, hindering scalability beyond a few qubits.