On the Policy Gradient Foundations of Group Relative Policy Optimization: Credit Assignment, Gradient Sparsity, and Rank Collapse
Group Relative Policy Optimization (GRPO) eliminates the learned critic in PPO by using the mean reward of grouped rollouts as a baseline. We provide a rigorous derivation of GRPO from first principles of the policy gradient theorem, revealing a fundamental credit assignment failure: under output-only reward, every token in a rollout receives identical advantage, collapsing token-level credit to a single scalar. We prove this induces gradient sparsity that intensifies over training, and demonstrate empirically via SVD analysis of GRPO gradients on Nemotron-4B/GSM8K that the gradient matrix has effective rank $\approx$ 2 regardless of group size $R \in \{2, 4, 8\}$. We formalize this as an intrinsic rank-2 structure arising from the zero-sum constraint on advantages and derive conditions under which GRPO's baseline is optimal. Our results characterize when GRPO's simplicity is theoretically justified and identify the credit assignment bottleneck as the key limitation for multi-step reasoning.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Demystifying Group Relative Policy Optimization: Its Policy Gradient is a U-Statistic
Group relative policy optimization (GRPO), a core methodological component of DeepSeekMath and DeepSeek-R1, has emerged as a cornerstone for scaling reasoning capabilities of large language models. Despite its widespread…
GUPO: Gradient Uncertainty-aware Policy Optimization for Post-Training Large Language Models
Group Relative Policy Optimization (GRPO) has become a widely used approach for post-training Large Language Models (LLMs) for reasoning. In GRPO, the group gradients induced by different queries within the same mini-bat…
GTPO: Stabilizing Group Relative Policy Optimization via Gradient and Entropy Control
Group Relative Policy Optimization (GRPO) is a promising policy-based approach for Large Language Model alignment, yet its performance is often limited by training instability and suboptimal convergence. In this paper, w…
DaGRPO: Rectifying Gradient Conflict in Reasoning via Distinctiveness-Aware Group Relative Policy Optimization
The evolution of Large Language Models (LLMs) has catalyzed a paradigm shift from superficial instruction following to rigorous long-horizon reasoning. While Group Relative Policy Optimization (GRPO) has emerged as a piv…
Mathematical ReasoningInstruction FollowingData AugmentationGVPO: Group Variance Policy Optimization for Large Language Model Post-Training
Post-training plays a crucial role in refining and aligning large language models to meet specific tasks and human preferences. While recent advancements in post-training techniques, such as Group Relative Policy Optimiz…
Language ModelingLanguage ModellingLarge Language Model