Annals of Mathematics and Artificial Intelligence最新文献

In this work, we study applications of the Conformal Prediction machine learning framework to the questions of statistical data testing. This technique is also known as Conformal Test Martingales. Earlier works on this topic used it to detect deviations from exchangeability assumptions (such as change points). Here we move to test popular hypergraphical models. We adopt and compare two versions of Conformal Testing Martingales. First: testing the data against exchangeability assumption, but using the elements of hypergraphical model for setting its parameters. Second: combining Conformal Testing Martingale with Hypergraphical On-Line Compression Models. The latter is an extension of the Conformal Prediction technique beyond exchangeability.

We show how these approaches help to accelerate the detection of data deviation from i.i.d. by making use of the knowledge about relations between the features embedded into a hypergraphical model.

Ensuring the safe and effective operation of autonomous systems is a complex undertaking that inherently relies on underlying decision-making processes. To rigorously analyze these processes, formal verification methods, such as model checking, offer a valuable means. However, the non-deterministic nature of realistic environments makes these approaches challenging and often impractical. This work explores the capabilities of a constraint-based planning approach, Tumato, in generating policies that guide the system to predefined goals while adhering to safety constraints. Constraint-based planning approaches are inherently able to provide guarantees of soundness and completeness. Our primary contribution lies in extending Tumato’s capabilities to accommodate non-deterministic outcomes of actions, enhancing the robustness of the behavior. Originally designed to accommodate only deterministic outcomes, actions can now be modeled to include alternative outcomes to address contingencies explicitly. The adapted solver generates policies that enable reaching the goals in a safe manner, even when such alternative outcomes of actions occur. Additionally, we introduce a purely declarative manner for specifying safety in Tumato to further enhance its expressiveness as well as to reduce the susceptibility to errors during specification. The incorporation of cost or duration values to actions enables the solver to restore safety in the most preferred manner when necessary. Finally, we highlight the overlap of Tumato’s safety-related capabilities with a systems-theoretic approach, STPA (Systems-Theoretic Process Analysis). The aim is to emphasize the ability to avoid unsafe control actions without their explicit identification, contributing to a more comprehensive and holistic understanding of safety.

Partially observable Markov decision processes (POMDP) are a useful model for decision-making under partial observability and stochastic actions. Partially Observable Monte-Carlo Planning (POMCP) is an online algorithm for deciding on the next action to perform, using a Monte-Carlo tree search approach, based on the UCT algorithm for fully observable Markov-decision processes. POMCP develops an action-observation tree, and at the leaves, uses a rollout policy to provide a value estimate for the leaf. As such, POMCP is highly dependent on the rollout policy to compute good estimates, and hence identify good actions. Thus, many practitioners who use POMCP are required to create strong, domain-specific heuristics. In this paper, we model POMDPs as stochastic contingent planning problems. This allows us to leverage domain-independent heuristics that were developed in the planning community. We suggest two heuristics, the first is based on the well-known (h_{add}) heuristic from classical planning, and the second is computed in belief space, taking the value of information into account.

Analogical inference is a remarkable capability of human reasoning, and has been used to solve hard reasoning tasks. Analogy based reasoning (AR) has gained increasing interest from the artificial intelligence community and has shown its potential in multiple machine learning tasks such as classification, decision making and recommendation with competitive results. We propose a deep learning (DL) framework to address and tackle two key tasks in AR: analogy detection and solving. The framework is thoroughly tested on the Siganalogies dataset of morphological analogical proportions (APs) between words, and shown to outperform symbolic approaches in many languages. Previous work have explored the behavior of the Analogy Neural Network for classification (ANNc) on analogy detection and of the Analogy Neural Network for retrieval (ANNr) on analogy solving by retrieval, as well as the potential of an autoencoder (AE) for analogy solving by generating the solution word. In this article we summarize these findings and we extend them by combining ANNr and the AE embedding model, and checking the performance of ANNc as an retrieval method. The combination of ANNr and AE outperforms the other approaches in almost all cases, and ANNc as a retrieval method achieves competitive or better performance than 3CosMul. We conclude with general guidelines on using our framework to tackle APs with DL.