The paper presents a model‑agnostic trust layer for machine‑learning surrogates of linear peeling‑ballooning stability, adding calibrated error bars, out‑of‑distribution alarms, and few‑shot recalibration. Experiments on JET‑like equilibria, BOUT++ surrogates, and SPARC/DIII‑D data show that the trust layer maintains nominal coverage, flags unseen cases, and restores accuracy with minimal additional simulations. The approach enables rapid, reliable stability screening for new fusion device designs.
arXiv:2607. 21407v1 Announce Type: cross Abstract: The boundary and divertor plasma govern how a tokamak exhausts power and particles, setting heat fluxes, target conditions, and the onset of detachment.
Fast and reliable plasma equilibrium prediction is essential for real-time tokamak operation and control, but conventional Grad-Shafranov (GS) solvers are often too costly for real-time deployment. We...
arXiv:2608.23217v1 Announce Type: cross Abstract: Fast and reliable plasma equilibrium prediction is essential for real-time tokamak operation and control, but conventional Grad-Shafranov (GS) solver...
arXiv:2606. 25151v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) embed governing equations in their loss function, enabling mesh-free solutions to partial differential equations.
arXiv:2606. 17529v1 Announce Type: cross Abstract: Scientific machine-learning (SciML) surrogates approximate expensive simulations, but exact expected outputs for arbitrary inputs are unavailable (the oracle problem).
The study introduces RegimeShift‑Surrogates, a streaming benchmark that tests surrogate models across eight tasks and multiple regimes. It compares revalidation—choosing the model with lowest current‑window validation loss—to stateful adaptive controllers and finds that revalidation consistently outperforms stateful methods, achieving lower mean log regret in most task‑scenario combinations. The results suggest that fresh validation evidence is more valuable than carrying over past evidence when dealing with distribution shifts.
arXiv:2608. 01378v1 Announce Type: new Abstract: Design campaigns in chemistry, materials science, and machine learning share a bottleneck: determining how good a candidate truly is requires an expensive evaluation - an experiment, a first-principles simulation, or a full training run.
The paper introduces a reference‑free diagnostic for physics‑informed neural networks that learn partial differential equation coefficients from noisy data. It presents three tests—an information‑matrix residual test, a heterogeneity statistic across window refits, and a Fisher‑rank statistic on the design—that can detect whether the operator used in the model is incorrect or whether the design is blind to certain rates, all without retraining the network. The authors demonstrate the method on a parabolic benchmark with 2 % noise, showing that the diagnostics reliably flag misspecified operators and blind designs while remaining silent when the model is correctly specified.
arXiv:2607. 00871v1 Announce Type: new Abstract: Self-evolving agents violate the assumption behind most learning-theoretic guarantees: the data, evaluator, components, and hypothesis space are produced by the policy being updated.
The paper introduces SWB-DM, a Byzantine‑robust federated learning aggregator that treats each slice of a client update as a one‑dimensional distribution, computes a trimmed Wasserstein barycenter across clients, and uses a medoid‑based gauge‑fixing step to recover coordinate identity. It further incorporates delayed‑momentum caching to decouple robustness from the specific clients sampled each round. Extensive experiments on CIFAR‑10, CIFAR‑100, FEMNIST, and a 500‑client scalability run reveal distinct failure modes of existing defenses and demonstrate that SWB‑DM achieves significant gains, especially when compared under equal round budgets.
arXiv:2608. 01130v1 Announce Type: new Abstract: A broad range of models face the mismatch where they are updated through trajectory losses but are evaluated by downstream task reward.
The paper presents a probabilistic neural surrogate for scrape‑off‑layer (SOL) plasma simulations in tokamaks. By mapping the curvilinear SOLPS‑ITER mesh to three fixed‑size image tensors, the authors preserve geometric adjacency and enable a convolutional network to process the mesh without loss of information. A conditional flow‑matching model is trained on this representation, producing efficient, scalable predictions that capture multiple plausible outcomes—such as distinct hot and cold modes—near the divertor detachment transition and correctly recover known bifurcations in synthetic data.