Search

Detecting anomalies in human-related videos is crucial for surveillance applications. Current methods primarily include appearance-based and action-based techniques. Appearance-based methods rely on low-level visual features such as color, texture, and shape. They learn a large number of pixel patterns and features related to known scenes during training, making them effective in detecting anomalies within these familiar contexts. However, when encountering new or significantly changed scenes, i.e., unknown scenes, they often fail because existing SOTA methods do not effectively capture the relationship between actions and their surrounding scenes, resulting in low generalization. In contrast, action-based methods focus on detecting anomalies in human actions but are usually less informative because they tend to overlook the relationship between actions and their scenes, leading to incorrect detection. For instance, the normal event of running on the beach and the abnormal event of running on the street might both be considered normal due to the lack of scene information. In short, current methods struggle to integrate low-level visual and high-level action features, leading to poor anomaly detection in varied and complex scenes. To address this challenge, we propose a novel decoupling-based architecture for human-related video anomaly detection (DecoAD). DecoAD significantly improves the integration of visual and action features through the decoupling and interweaving of scenes and actions, thereby enabling a more intuitive and accurate understanding of complex behaviors and scenes. DecoAD supports fully supervised, weakly supervised, and unsupervised settings., Comment: 13pages, 9 figures

Multi-behavior recommendation systems enhance effectiveness by leveraging auxiliary behaviors (such as page views and favorites) to address the limitations of traditional models that depend solely on sparse target behaviors like purchases. Existing approaches to multi-behavior recommendations typically follow one of two strategies: some derive initial node representations from individual behavior subgraphs before integrating them for a comprehensive profile, while others interpret multi-behavior data as a heterogeneous graph, applying graph neural networks to achieve a unified node representation. However, these methods do not adequately explore the intricate patterns of behavior among users and items. To bridge this gap, we introduce a novel algorithm called Behavior Pattern mining-based Multi-behavior Recommendation (BPMR). Our method extensively investigates the diverse interaction patterns between users and items, utilizing these patterns as features for making recommendations. We employ a Bayesian approach to streamline the recommendation process, effectively circumventing the challenges posed by graph neural network algorithms, such as the inability to accurately capture user preferences due to over-smoothing. Our experimental evaluation on three real-world datasets demonstrates that BPMR significantly outperforms existing state-of-the-art algorithms, showing an average improvement of 268.29% in Recall@10 and 248.02% in NDCG@10 metrics. The code of our BPMR is openly accessible for use and further research at https://github.com/rookitkitlee/BPMR.

Graph learning models have been widely deployed in collaborative filtering (CF) based recommendation systems. Due to the issue of data sparsity, the graph structure of the original input lacks potential positive preference edges, which significantly reduces the performance of recommendations. In this paper, we study how to enhance the graph structure for CF more effectively, thereby optimizing the representation of graph nodes. Previous works introduced matrix completion techniques into CF, proposing the use of either stochastic completion methods or superficial structure completion to address this issue. However, most of these approaches employ random numerical filling that lack control over noise perturbations and limit the in-depth exploration of higher-order interaction features of nodes, resulting in biased graph representations. In this paper, we propose an Amplify Graph Learning framework based on Sparsity Completion (called AGL-SC). First, we utilize graph neural network to mine direct interaction features between user and item nodes, which are used as the inputs of the encoder. Second, we design a factorization-based method to mine higher-order interaction features. These features serve as perturbation factors in the latent space of the hidden layer to facilitate generative enhancement. Finally, by employing the variational inference, the above multi-order features are integrated to implement the completion and enhancement of missing graph structures. We conducted benchmark and strategy experiments on four real-world datasets related to recommendation tasks. The experimental results demonstrate that AGL-SC significantly outperforms the state-of-the-art methods.

Explicit mathematical reconstructions of quantum networks play a significant role in developing quantum information science. However, tremendous parameter requirements and physical constraint implementations have become computationally non-ignorable encumbrances. In this work, we propose an efficient method for quantum network tomography by learning isometries on the Stiefel manifold. Tasks of reconstructing quantum networks are tackled by solving a series of unconstrained optimization problems with significantly fewer parameters. The stepwise isometry estimation shows the capability for providing information of the truncated quantum network while processing the tomography. Remarkably, this method enables the dimension-reduced quantum network tomography by reducing the ancillary dimensions of isometries with bounded error. As a result, our proposed method exhibits high accuracy and efficiency.

Reinforcement Learning from AI Feedback (RLAIF) has the advantages of shorter annotation cycles and lower costs over Reinforcement Learning from Human Feedback (RLHF), making it highly efficient during the rapid strategy iteration periods of large language model (LLM) training. Using ChatGPT as a labeler to provide feedback on open-domain prompts in RLAIF training, we observe an increase in human evaluators' preference win ratio for model responses, but a decrease in evaluators' satisfaction rate. Analysis suggests that the decrease in satisfaction rate is mainly due to some responses becoming less helpful, particularly in terms of correctness and truthfulness, highlighting practical limitations of basic RLAIF. In this paper, we propose Hybrid Reinforcement Learning from AI Feedback (HRLAIF). This method enhances the accuracy of AI annotations for responses, making the model's helpfulness more robust in training process. Additionally, it employs AI for Red Teaming, further improving the model's harmlessness. Human evaluation results show that HRLAIF inherits the ability of RLAIF to enhance human preference for outcomes at a low cost while also improving the satisfaction rate of responses. Compared to the policy model before Reinforcement Learning (RL), it achieves an increase of 2.08\% in satisfaction rate, effectively addressing the issue of a decrease of 4.58\% in satisfaction rate after basic RLAIF., Comment: 18 pages, 7 figures

Engineering quantum devices requires reliable characterization of the quantum system, including qubits, quantum operations (also known as instruments) and the quantum noise. Recently, quantum gate set tomography (GST) has emerged as a powerful technique for self-consistently describing quantum states, gates, and measurements. However, non-Markovian correlations between the quantum system and environment impact the reliability of GST. To address this, we propose a self-consistent operational framework called instrument set tomography (IST) for non-Markovian GST. Based on the stochastic quantum process, the instrument set describes instruments and system-environment (SE) correlations. We introduce a linear inversion IST (LIST) to describe instruments and SE correlations without physical constraints. The disharmony of linear relationships between instruments is detected. Furthermore, we propose a physically constrained statistical method based on the maximum likelihood estimation for IST (MLE-IST) with adjustable dimensions. MLE-IST shows significant flexibility in adapting to different types of devices, such as noisy intermediate-scale quantum (NISQ) devices, by adjusting the model and constraints. Experimental results demonstrate the effectiveness and necessity of simultaneously describing instruments and SE correlations. Remarkably, real-chip experiments indicate that a polynomial number of parameters with respect to the Markovian order are sufficient to characterize non-Markovian quantum noise in current NISQ devices. Consequently, IST provides an essential and self-consistent framework for characterizing, benchmarking, and developing quantum devices in terms of the instrument set., Comment: Quantum Science and Technology 2024

Variational quantum eigensolver (VQE), aiming at determining the ground state energy of a quantum system described by a Hamiltonian on noisy intermediate scale quantum (NISQ) devices, is among the most significant applications of variational quantum algorithms (VQAs). However, the accuracy and trainability of the current VQE algorithm are significantly influenced due to the barren plateau (BP), the non-negligible gate error and limited coherence time in NISQ devices. To tackle these issues, a gradient-sensitive alternate framework with variable ansatz is proposed in this paper to enhance the performance of the VQE. We first propose a theoretical framework for VA-VQE via alternately solving a multi-objective optimization problem and the original VQE, where the multi-objective optimization problem is defined with respect to cost function values and gradient magnitudes. Then, we propose a novel implementation method based on the double $\epsilon$-greedy strategy with the candidate tree and modified multi-objective genetic algorithm. As a result, the local optima are avoided both in ansatz and parameter perspectives, and the stability of output ansatz is enhanced. The experimental results indicate that our framework shows considerably improvement of the error of the found solution by up to 87.9% compared with the hardware-efficient ansatz. Furthermore, compared with the full-randomized VA-VQE implementation, our framework is able to obtain the improvement of the error and the stability by up to 36.0% and 58.7%, respectively, with similar quantum costs., Comment: 30 pages, 17 figures