Bayesian experimental design (BED) has traditionally been based on maximising expected uncertainty reductions from prior to posterior. A major shortfall of this approach is that it leads to doubly intractable objectives that are difficult to optimise, while customising them to particular downstream tasks of interest can also be difficult. Following first principles decision theory, we demonstrate that BED can alternatively be formulated in terms of an expected future loss (EFL) on downstream actions, providing a simple and naturally task-driven framework. Critically, we then show that all such EFLs can be rearranged into singly intractable objectives that can be jointly optimised with respect to both the design policy and a downstream action policy using stochastic gradients, an approach we refer to as ACTION-BED. This formulation further sidesteps the need for any explicit posterior or marginal likelihood estimation and is naturally implicit, requiring only the ability to sample from the joint model over model parameters and data, and evaluate the downstream loss function. It thus allows design policies to be learned more effectively, efficiently, and simply than existing methods, while providing easy customisation to different downstream tasks and losses.

When this structure is known, at least locally, it can be exploited through reduction mappings that reparametrise part of the parameter space to lie on the solution manifold. These reductions naturally arise from inner optimisation problems and effectively remove redundant directions, yielding a lowerdimensional objective. In this work, we introduce a general framework to understand how such reductions influence the optimisation landscape. We show that well-designed reduction mappings improve curvature properties of the objective, leading to better-conditioned problems and theoretically faster convergence for gradient-based methods. Our analysis unifies a range of scenarios where structural information at optimality is leveraged to accelerate convergence, offering a principled explanation for the empirical gains observed in such optimisation algorithms.

Vision-Language Latent Diffusion Models (LDMs) (Rombach et al., 2022) provide powerful generative priors for inverse problems. However, existing LDM-based inverse solvers typically require a large number of neural function evaluations (NFEs) and backpropagation through large pretrained components, leading to substantial computational costs and, in some cases, degraded reconstruction quality. We propose a unified Euclidean-Wasserstein-2 gradient-flow framework that jointly performs posterior sampling and prompt optimization in the latent space through a single flow that aligns the prior and posterior with the observed data. Combined with few-step latent text-to-image models, this formulation enables low-NFE inference without backpropagation through autoencoders. Experiments across several canonical imaging inverse problems show that our method achieves state-of-the-art performance with significantly reduced computational cost.

By minimising off-target activation, Bayesian target optimisation could enable (e.g.)421 more precise synaptic connectivity mapping, improving our understanding of neural circuitry. This422 advancement has potential implications for understanding brain disorders like epilepsy, where423 abnormal synaptic connections are central to seizure generation and propagation. Deepening our424 understanding of these diseases can lead to enhanced targeted interventions and more effective425 therapeutic strategies, benefiting individuals with neurological disorders.426 First, we develop431 the approach for single optogenetic targets, as this is most closely related to existing GP-based432 receptive field inference techniques. We use a GP-Bernoulli approach to model the response ynt of436 neuron n on trial t to a single-target stimulus xt,437 ynt Bernoulli( (gn(xt))), (9) where the stimulus xt =( c1t,c2t,It) 2 R3 represents the two-dimensional coordinates and laser438 power of the t-th hologram.

Now consider a perturbation of the prior distribution over transition functions δ: T R 0 such that R Tp δ(Tp)P(Tp|h0)dTp = 1. Proof: Proposition 2 directly extends Proposition 1 in [8] to BAMDPs. Therefore, the perturbed distribution over histories is also a valid probability distribution. Provided that cbo is chosen appropriately (details in the appendix), as the number of perturbations expanded approaches, a perturbation within any > 0 of the optimal perturbation will be expanded by the Bayesian optimisation procedure with probability 1 δ. Proof: Consider an adversary decision node, v, associated with augmented state (s,ha,y) in the BACVaR-SG. We begin by proving that Q((s,ha,y),ξ) is continuous with respect to ξ. Define a function d: S R, such that ξ + d produces a valid adversary perturbation.

In this work, we address risk-averse Bayes-adaptive reinforcement learning. We pose the problem of optimising the conditional value at risk (CVaR) of the total return in Bayes-adaptive Markov decision processes (MDPs). We show that a policy optimising CVaR in this setting is risk-averse to both the epistemic uncertainty due to the prior distribution over MDPs, and the aleatoric uncertainty due to the inherent stochasticity of MDPs. We reformulate the problem as a two-player stochastic game and propose an approximate algorithm based on Monte Carlo tree search and Bayesian optimisation. Our experiments demonstrate that our approach significantly outperforms baseline approaches for this problem.