We outline how to create a mechanism that provides an optimal way to elicit,
from an arbitrary group of experts, the probability of the truth of an
arbitrary logical proposition together with collective information that has an
explicit form and interprets this probability. Namely, we provide strong
arguments for the possibility of the development of a self-resolving prediction
market with play money that incentivizes direct information exchange between
experts. Such a system could, in particular, motivate experts from all over the
world to collectively solve scientific or medical problems in a very efficient
manner. In our main considerations about real experts, they are not assumed to
be Bayesian and their behavior is described by utilities that satisfy the von
Neumann--Morgenstern axioms only locally.

提出一种使用预测市场机制，通过激励专家之间直接信息交流的方式来解决科学或医学领域问题的方法。

非理性人类智能的可解释集体智能

Interpretable collective intelligence of non-rational human agents

Machine learning lies at the heart of new possibilities for scientific
discovery, knowledge generation, and artificial intelligence. Its potential
benefits to these fields requires going beyond predictive accuracy and focusing
on interpretability. In particular, many scientific problems require
interpretations in a domain-specific interpretable feature space (e.g. the
frequency domain) whereas attributions to the raw features (e.g. the pixel
space) may be unintelligible or even misleading. To address this challenge, we
propose TRIM (TRansformation IMportance), a novel approach which attributes
importances to features in a transformed space and can be applied post-hoc to a
fully trained model. TRIM is motivated by a cosmological parameter estimation
problem using deep neural networks (DNNs) on simulated data, but it is
generally applicable across domains/models and can be combined with any local
interpretation method. In our cosmology example, combining TRIM with contextual
decomposition shows promising results for identifying which frequencies a DNN
uses, helping cosmologists to understand and validate that the model learns
appropriate physical features rather than simulation artifacts.

提出了 TRIM（TRansformation IMportance）方法，该方法可将特征权重归因于转换空间，可用于任何领域 / 模型，并可与任何局部解释方法相结合。在宇宙学参数估计问题中，将 TRIM 与上下文分解相结合，有助于识别 DNN 使用的频率，以了解模型是否学习了适当的物理特征，而不是模拟工件。

应用于宇宙学的变换重要性

Transformation Importance with Applications to Cosmology

In addition to providing high-profile successes in computer vision and
natural language processing, neural networks also provide an emerging set of
techniques for scientific problems. Such data-driven models, however, typically
ignore physical insights from the scientific system under consideration. Among
other things, a physics-informed model formulation should encode some degree of
stability or robustness or well-conditioning (in that a small change of the
input will not lead to drastic changes in the output), characteristic of the
underlying scientific problem. We investigate whether it is possible to include
physics-informed prior knowledge for improving the model quality (e.g.,
generalization performance, sensitivity to parameter tuning, or robustness in
the presence of noisy data). To that extent, we focus on the stability of an
equilibrium, one of the most basic properties a dynamic system can have, via
the lens of Lyapunov analysis. For the prototypical problem of fluid flow
prediction, we show that models preserving Lyapunov stability improve the
generalization error and reduce the prediction uncertainty.

该论文探讨了如何将物理先验知识加入数据驱动模型，提高模型的质量和稳定性，并以 Lyapunov 分析为视角，通过研究液体流动预测的典型问题，验证了保持 Lyapunov 稳定性的模型可以提高泛化性能并减少预测不确定性。