The paper introduces Decomposition-Aware Distributional Optimization (DADO), a new algorithm that exploits decomposability in property predictors to improve in‑silico design of discrete objects such as proteins, circuits, and materials. DADO uses a soft‑factorized search distribution and graph message‑passing to coordinate optimization across linked factors defined by a junction tree over design variables. The method aims to make distributional optimization over combinatorial design spaces more efficient by leveraging the structure of the predictive model.
arXiv:2607. 19519v1 Announce Type: cross Abstract: Most 3D properties relevant to molecular design, including free energies and shape descriptors, are $\textit{expectations}$ over the Boltzmann distribution over 3D configurations of a molecular graph.
arXiv:2602. 15648v2 Announce Type: replace Abstract: Inverse design problems are common in engineering and materials science.
arXiv:2610.22763v1 Announce Type: new Abstract: Bayesian optimization (BO) over discrete spaces is challenging due to missing gradients and the sparsity of valid solutions. Latent-space BO sidesteps...
arXiv:2607. 23480v1 Announce Type: new Abstract: Variational autoencoders (VAEs) transform high-dimensional, often noisy data into a compact latent representation, making downstream optimization more tractable.
SABLE is a Bayesian optimisation method designed for high‑dimensional latent spaces where the latent dimension far exceeds the evaluation budget. It uses a Bayesian affine surrogate that decomposes each query into an in‑span component optimized by expected improvement and an off‑span component governed by a scalar norm, with a finite‑horizon schedule to avoid under‑exploration. A recompression scheme reduces the number of pseudo‑observations to ≠O(M), keeping refitting costs manageable while achieving high sample efficiency and rapid decision times on synthetic and latent‑space optimisation tasks.
arXiv:2609. 22752v1 Announce Type: new Abstract: Diffusion models have demonstrated strong power in generative modeling tasks across multiple domains, exhibiting a remarkable capability of learning complex distributions from samples.
arXiv:2508. 20330v5 Announce Type: replace Abstract: Combinatorial optimization problems are ubiquitous in science and engineering.
arXiv:2608. 15669v1 Announce Type: new Abstract: Scientific discovery often involves optimising expensive-to-evaluate objectives over vast, structured, and open-ended hypothesis spaces, such as molecules, protein sequences, and computer programs.
arXiv:2602. 10392v2 Announce Type: replace Abstract: When designing new materials, it is often necessary to tailor the material design to have some desired properties.
DisCoMBO introduces the Distributional Conformance Score (DisCo), a bounded metric that blends the flexibility of Probabilistic Circuits with formal uncertainty quantification for Sequential Model-Based Optimization. DisCo measures model surprise in a [0,1] range, retaining linear-time inference while enabling accurate assessment of external knowledge conformance. The resulting DisCoMBO framework is proven to be a zero-regret algorithm and shows strong performance on benchmarks in AutoML, material optimization, and wind park optimization.
arXiv:2606. 08802v1 Announce Type: new Abstract: Standard flow and diffusion pre-training matches the distribution of available data (e.
arXiv:2610.08367v1 Announce Type: new Abstract: Several discrete design tasks, such as molecular discovery, require diverse collections of useful candidates at low computational cost. High validity a...