{"slug": "emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement", "title": "Emergent Sparsity in Frozen Random CNN Feature Extractors for Deep Reinforcement Learning", "summary": "A new arXiv preprint (2607.26059v1) reports that deep reinforcement learning agents using frozen, randomly initialized CNN feature extractors spontaneously develop extremely sparse fully-connected representations, with as few as 1-3 active neurons out of 64 for deterministic Pong, while trainable CNNs activate 55-64 neurons. The study finds that sparsity scales with task complexity, that within-game seed variation can bound performance, and that ablating the active neurons crashes performance across two PPO implementations and four games.", "body_md": "arXiv:2607.26059v1 Announce Type: new\nAbstract: We report a striking phenomenon: deep reinforcement learning agents trained with frozen, randomly initialized CNN feature extractors spontaneously develop extremely sparse fully-connected representations, without any sparsity-inducing objective. In the first fully-connected layer (FC1, $3{,}136 \\to 64$), agents compress task-relevant information through as few as 1-3 neurons out of 64 for deterministic Pong (5-11 for stochastic Pong), while trainable CNNs activate 55-64 neurons under matched conditions. We establish four principal findings. First, FC1 sparsity scales with task complexity: 1-11 for Pong, 19-26 for Breakout, and $\\sim$42 for Space Invaders. Width-scaling confirms this reflects task structure rather than a fixed capacity fraction. Second, within-game scaling emerges: three identical Pong seeds produce 5, 7, and 11 active neurons. The 5-neuron seed plateaus at $+14$ reward, while the others reach expert performance ($+18.4$, $+18.7$), suggesting the random projection's usable dimensionality bounds achievable performance. Third, ablation confirms necessity: removing these active neurons crashes performance across two PPO implementations and four games. Fourth, the information bottleneck commits early: a sweep shows the active set locks by 15-30M steps, while reward turns positive 35-105M steps later. A complementary finding in Breakout shows frozen and trainable CNNs reach competitive rewards via structurally different bottlenecks: frozen agents use 17-25 active neurons (participation ratio $\\sim$10-14), while trainable agents use 51 (participation ratio $\\sim$3.6). Finally, wherever input dimensionality dwarfs intrinsic task dimensionality, gradient descent on a frozen random projection may reveal the effective rank of the underlying problem without explicit sparsity machinery.", "url": "https://wpnews.pro/news/emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement", "canonical_source": "https://arxiv.org/abs/2607.26059", "published_at": "2026-07-30 04:00:00+00:00", "updated_at": "2026-07-30 04:29:55.885668+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "neural-networks"], "entities": ["arXiv"], "alternates": {"html": "https://wpnews.pro/news/emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement", "markdown": "https://wpnews.pro/news/emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement.md", "text": "https://wpnews.pro/news/emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement.txt", "jsonld": "https://wpnews.pro/news/emergent-sparsity-in-frozen-random-cnn-feature-extractors-for-deep-reinforcement.jsonld"}}