{"slug": "rethinking-learning-based-influence-maximization-simple-neural-surrogates-and", "title": "Rethinking Learning-Based Influence Maximization: Simple Neural Surrogates and Native Discrete Search", "summary": "Researchers propose SIMBA, a diffusion-model-agnostic framework for influence maximization that replaces complex neural architectures with a lightweight two-layer graph neural network surrogate and direct discrete search via batched multi-swap simulated annealing. The method reduces time-to-solution while improving influence spread and data efficiency, with code available on GitHub.", "body_md": "arXiv:2608.08406v1 Announce Type: new\nAbstract: Existing learning-based influence maximization frameworks rely heavily on complex neural architectures and continuous optimization over seed representations. We challenge this paradigm with SIMBA, a diffusion-model-agnostic framework pairing a lightweight neural surrogate with direct discrete search. SIMBA introduces three key components: 1) uniformly anchored node embeddings that eliminate initialization noise and encourage learning driven by graph topology and diffusion pattern, 2) a shallow two-layer graph neural network surrogate predicting final infection states, and 3) batched multi-swap simulated annealing that explores combinatorial seed space without gradients or continuous relaxation. By shifting compute from complex representation learning to effective discrete search, SIMBA drastically cuts time-to-solution while achieving superior influence spread and data efficiency. Our code is available at https://github.com/yl489/rethink-IM.", "url": "https://wpnews.pro/news/rethinking-learning-based-influence-maximization-simple-neural-surrogates-and", "canonical_source": "https://www.machinebrief.com/news/rethinking-learning-based-influence-maximization-simple-neur-ygm9", "published_at": "2026-08-11 04:00:00+00:00", "updated_at": "2026-08-11 05:14:03.472508+00:00", "lang": "en", "topics": ["machine-learning", "neural-networks", "artificial-intelligence"], "entities": ["SIMBA", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/rethinking-learning-based-influence-maximization-simple-neural-surrogates-and", "markdown": "https://wpnews.pro/news/rethinking-learning-based-influence-maximization-simple-neural-surrogates-and.md", "text": "https://wpnews.pro/news/rethinking-learning-based-influence-maximization-simple-neural-surrogates-and.txt", "jsonld": "https://wpnews.pro/news/rethinking-learning-based-influence-maximization-simple-neural-surrogates-and.jsonld"}}