We consider discounted infinite horizon constrained Markov decision processes (CMDPs) where the goal is to find an optimal policy that maximizes the expected cumulative reward subject to expected cumulative constraints. Motivated by the application of CMDPs in online learning of safety-critical systems, we focus on developing an algorithm that ensures constraint satisfaction during learning. To this end, we develop a zeroth-order interior point approach based on the log barrier function of the CMDP. Under the commonly assumed conditions of Fisher non-degeneracy and bounded transfer error of the policy parameterization, we establish the theoretical properties of the algorithm. In particular, in contrast to existing CMDP approaches that ensure policy feasibility only upon convergence, our algorithm guarantees feasibility of the policies during the learning process and converges to the optimal policy with a sample complexity of $O(\varepsilon^{-6})$. In comparison to the state-of-the-art policy gradient-based algorithm, C-NPG-PDA, our algorithm requires an additional $O(\varepsilon^{-2})$ samples to ensure policy feasibility during learning with same Fisher-non-degenerate parameterization.
相關內容
Traffic prediction, a critical component for intelligent transportation systems, endeavors to foresee future traffic at specific locations using historical data. Although existing traffic prediction models often emphasize developing complex neural network structures, their accuracy has not seen improvements accordingly. Recently, Large Language Models (LLMs) have shown outstanding capabilities in time series analysis. Differing from existing models, LLMs progress mainly through parameter expansion and extensive pre-training while maintaining their fundamental structures. In this paper, we propose a Spatial-Temporal Large Language Model (ST-LLM) for traffic prediction. Specifically, ST-LLM redefines the timesteps at each location as tokens and incorporates a spatial-temporal embedding module to learn the spatial location and global temporal representations of tokens. Then these representations are fused to provide each token with unified spatial and temporal information. Furthermore, we propose a novel partially frozen attention strategy of the LLM, which is designed to capture spatial-temporal dependencies for traffic prediction. Comprehensive experiments on real traffic datasets offer evidence that ST-LLM outperforms state-of-the-art models. Notably, the ST-LLM also exhibits robust performance in both few-shot and zero-shot prediction scenarios.
In stereo-matching knowledge distillation methods of the self-supervised monocular depth estimation, the stereo-matching network's knowledge is distilled into a monocular depth network through pseudo-depth maps. In these methods, the learning-based stereo-confidence network is generally utilized to identify errors in the pseudo-depth maps to prevent transferring the errors. However, the learning-based stereo-confidence networks should be trained with ground truth (GT), which is not feasible in a self-supervised setting. In this paper, we propose a method to identify and filter errors in the pseudo-depth map using multiple disparity maps by checking their consistency without the need for GT and a training process. Experimental results show that the proposed method outperforms the previous methods and works well on various configurations by filtering out erroneous areas where the stereo-matching is vulnerable, especially such as textureless regions, occlusion boundaries, and reflective surfaces.
In the context of interactive proofs, a "folding scheme" (popularized by Nova) is a way to combine multiple instances of a constraint system into a single instance, so the validity of the multiple instances can statistically be reduced to the validity of a single one. We show how Nova folding can be generalized to ``custom'' gates and extra rounds of verifier randomness. As an application of this extension, we present Origami, the first (to our knowledge) known example of a folding scheme for lookups.
We study the complexity of constructive bribery in the context of structured multiwinner approval elections. Given such an election, we ask whether a certain candidate can join the winning committee by adding, deleting, or swapping approvals, where each such action comes at a cost and we are limited by a budget. We assume our elections to either have the candidate interval or the voter interval property, and we require the property to hold also after the bribery. While structured elections usually make manipulative attacks significantly easier, our work also shows examples of the opposite behavior. We conclude by presenting preliminary insights regarding the destructive variant of our problem.
Encompassing numerous nationwide, statewide, and institutional initiatives in the United States, provider profiling has evolved into a major health care undertaking with ubiquitous applications, profound implications, and high-stakes consequences. In line with such a significant profile, the literature has accumulated a number of developments dedicated to enhancing the statistical paradigm of provider profiling. Tackling wide-ranging profiling issues, these methods typically adjust for risk factors using linear predictors. While this approach is simple, it can be too restrictive to characterize complex and dynamic factor-outcome associations in certain contexts. One such example arises from evaluating dialysis facilities treating Medicare beneficiaries with end-stage renal disease. It is of primary interest to consider how the coronavirus disease (COVID-19) affected 30-day unplanned readmissions in 2020. The impact of COVID-19 on the risk of readmission varied dramatically across pandemic phases. To efficiently capture the variation while profiling facilities, we develop a generalized partially linear model (GPLM) that incorporates a neural network. Considering provider-level clustering, we implement the GPLM as a stratified sampling-based stochastic optimization algorithm that features accelerated convergence. Furthermore, an exact test is designed to identify under- and over-performing facilities, with an accompanying funnel plot to visualize profiles. The advantages of the proposed methods are demonstrated through simulation experiments and profiling dialysis facilities using 2020 Medicare claims from the United States Renal Data System.
Many real-world problems can be efficiently modeled as Mixed Integer Programs (MIPs) and solved with the Branch-and-Bound method. Prior work has shown the existence of MIP backdoors, small sets of variables such that prioritizing branching on them when possible leads to faster running times. However, finding high-quality backdoors that improve running times remains an open question. Previous work learns to estimate the relative solver speed of randomly sampled backdoors through ranking and then decide whether to use it. In this paper, we utilize the Monte-Carlo tree search method to collect backdoors for training, rather than relying on random sampling, and adapt a contrastive learning framework to train a Graph Attention Network model to predict backdoors. Our method, evaluated on four common MIP problem domains, demonstrates performance improvements over both Gurobi and previous models.
Graph Neural Networks (GNNs) have recently become increasingly popular due to their ability to learn complex systems of relations or interactions arising in a broad spectrum of problems ranging from biology and particle physics to social networks and recommendation systems. Despite the plethora of different models for deep learning on graphs, few approaches have been proposed thus far for dealing with graphs that present some sort of dynamic nature (e.g. evolving features or connectivity over time). In this paper, we present Temporal Graph Networks (TGNs), a generic, efficient framework for deep learning on dynamic graphs represented as sequences of timed events. Thanks to a novel combination of memory modules and graph-based operators, TGNs are able to significantly outperform previous approaches being at the same time more computationally efficient. We furthermore show that several previous models for learning on dynamic graphs can be cast as specific instances of our framework. We perform a detailed ablation study of different components of our framework and devise the best configuration that achieves state-of-the-art performance on several transductive and inductive prediction tasks for dynamic graphs.
Multi-relation Question Answering is a challenging task, due to the requirement of elaborated analysis on questions and reasoning over multiple fact triples in knowledge base. In this paper, we present a novel model called Interpretable Reasoning Network that employs an interpretable, hop-by-hop reasoning process for question answering. The model dynamically decides which part of an input question should be analyzed at each hop; predicts a relation that corresponds to the current parsed results; utilizes the predicted relation to update the question representation and the state of the reasoning process; and then drives the next-hop reasoning. Experiments show that our model yields state-of-the-art results on two datasets. More interestingly, the model can offer traceable and observable intermediate predictions for reasoning analysis and failure diagnosis, thereby allowing manual manipulation in predicting the final answer.
Detecting carried objects is one of the requirements for developing systems to reason about activities involving people and objects. We present an approach to detect carried objects from a single video frame with a novel method that incorporates features from multiple scales. Initially, a foreground mask in a video frame is segmented into multi-scale superpixels. Then the human-like regions in the segmented area are identified by matching a set of extracted features from superpixels against learned features in a codebook. A carried object probability map is generated using the complement of the matching probabilities of superpixels to human-like regions and background information. A group of superpixels with high carried object probability and strong edge support is then merged to obtain the shape of the carried object. We applied our method to two challenging datasets, and results show that our method is competitive with or better than the state-of-the-art.
Recommender System (RS) is a hot area where artificial intelligence (AI) techniques can be effectively applied to improve performance. Since the well-known Netflix Challenge, collaborative filtering (CF) has become the most popular and effective recommendation method. Despite their success in CF, various AI techniques still have to face the data sparsity and cold start problems. Previous works tried to solve these two problems by utilizing auxiliary information, such as social connections among users and meta-data of items. However, they process different types of information separately, leading to information loss. In this work, we propose to utilize Heterogeneous Information Network (HIN), which is a natural and general representation of different types of data, to enhance CF-based recommending methods. HIN-based recommender systems face two problems: how to represent high-level semantics for recommendation and how to fuse the heterogeneous information to recommend. To address these problems, we propose to applying meta-graph to HIN-based RS and solve the information fusion problem with a "matrix factorization (MF) + factorization machine (FM)" framework. For the "MF" part, we obtain user-item similarity matrices from each meta-graph and adopt low-rank matrix approximation to get latent features for both users and items. For the "FM" part, we propose to apply FM with Group lasso (FMG) on the obtained features to simultaneously predict missing ratings and select useful meta-graphs. Experimental results on two large real-world datasets, i.e., Amazon and Yelp, show that our proposed approach is better than that of the state-of-the-art FM and other HIN-based recommending methods.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191