1 citations · 3 across the 22 of their papers we have counts for
8 papers · 1 filter
Overcoming Forgetting in LLM Fine-Tuning with Evolution Strategies
Kajetan Schweighofer, Conor F. Hayes, Roberto Dailey +2
Evolution Strategies (ES) has recently emerged as a competitive alternative to reinforcement learning (RL) for large language model (LLM) fine-tuning, offering advantages through s…
Efficient Pre-Training of LLMs through Truncated SVD Representations
Kaivan Kamali, Kajetan Schweighofer, Hormoz Shahrzad +3
LLM pretraining is extremely costly; therefore, parameter-efficient LLM architectures have recently emerged as a compelling research direction. One such promising approach is to re…
Optimize Wider, Not Deeper: Consensus Aggregation for Policy Optimization
Zelal Su, Mustafaoglu, Sungyoung Lee +3
Proximal policy optimization (PPO) approximates the trust region update using multiple epochs of clipped SGD. Each epoch may drift further from the natural gradient direction, crea…
Quantized Evolution Strategies: High-precision Fine-tuning of Quantized LLMs at Low-precision Cost
Yinggan Xu, Kajetan Schweighofer, Risto Miikkulainen +1
Post-Training Quantization (PTQ) is essential for deploying Large Language Models (LLMs) on memory-constrained devices, yet it renders models static and difficult to fine-tune. Sta…
The Blessing of Dimensionality in LLM Fine-tuning: A Variance-Curvature Perspective
Qiyao Liang, Jinyeop Song, Yizhou Liu +4
Weight-perturbation evolution strategies (ES) can fine-tune billion-parameter language models with surprisingly small populations (e.g., ), contradicting classical…
Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning
Xin Qiu, Yulu Gan, Conor F. Hayes +6
Fine-tuning large language models (LLMs) for downstream tasks is an essential stage of modern AI deployment. Reinforcement learning (RL) has emerged as the dominant fine-tuning par…
Evolutionary Policy Optimization
Zelal Su "Lain" Mustafaoglu, Keshav Pingali, Risto Miikkulainen
A key challenge in reinforcement learning (RL) is managing the exploration-exploitation trade-off without sacrificing sample efficiency. Policy gradient (PG) methods excel in explo…
Effective Regularization Through Loss-Function Metalearning
Santiago Gonzalez, Xin Qiu, Risto Miikkulainen
Evolutionary computation can be used to optimize several different aspects of neural network architectures. For instance, the TaylorGLO method discovers novel, customized loss func…