EPISODE · Nov 25, 2025 · 14 MIN
Evolution Strategies at the Hyperscale
from Best AI papers explained · host Enoch H. Kang
This paper introduces Evolution Guided General Optimization via Low-rank Learning (EGGROLL), a novel algorithm that enhances the scalability of **Evolution Strategies (ES)** for optimizing neural networks with billions of parameters. ES is an optimization method that bypasses the need for gradient backpropagation, offering advantages like handling non-differentiable objectives and superior parallelization potential. EGGROLL overcomes the memory and computational bottlenecks of traditional ES by substituting expensive full-rank parameter perturbations with **efficient low-rank matrix perturbations**, significantly increasing training throughput, as demonstrated by a **hundredfold speedup** for large models. The architecture is tested across various domains, including reinforcement learning, large language model (LLM) fine-tuning for reasoning tasks, and the stable pretraining of a custom **purely integer-based recurrent language model (EGG)**, showcasing the method's efficiency and flexibility. A **theoretical analysis** also confirms that the low-rank update rapidly converges to the full-rank ES update.
What this episode covers
This paper introduces Evolution Guided General Optimization via Low-rank Learning (EGGROLL), a novel algorithm that enhances the scalability of **Evolution Strategies (ES)** for optimizing neural networks with billions of parameters. ES is an optimization method that bypasses the need for gradient backpropagation, offering advantages like handling non-differentiable objectives and superior parallelization potential. EGGROLL overcomes the memory and computational bottlenecks of traditional ES by substituting expensive full-rank parameter perturbations with **efficient low-rank matrix perturbations**, significantly increasing training throughput, as demonstrated by a **hundredfold speedup** for large models. The architecture is tested across various domains, including reinforcement learning, large language model (LLM) fine-tuning for reasoning tasks, and the stable pretraining of a custom **purely integer-based recurrent language model (EGG)**, showcasing the method's efficiency and flexibility. A **theoretical analysis** also confirms that the low-rank update rapidly converges to the full-rank ES update.
NOW PLAYING
Evolution Strategies at the Hyperscale
No transcript for this episode yet
Similar Episodes
Mar 31, 2026 ·54m
Mar 27, 2026 ·14m
Mar 24, 2026 ·42m
Mar 20, 2026 ·42m
Mar 17, 2026 ·41m
Mar 13, 2026 ·44m