In the era of intelligent computing, with the aid of Internet of Things (IoT) technology, artificial intelligence (AI) chips can be embedded at the terminal, object, edge, and cloud levels, ultimately achieving the vision where there is computation, there is AI intelligence. This not only enhances the efficiency of production and daily life but also exponentially increases the risk of privacy information leakage within AI models. This article leverages the characteristics of strong physical unclonable functions (PUFs), in which the inherent feature information is hidden in physical variations and difficult to steal, to design a security protocol based on strong PUFs that provides effective protection for AI model parameters in the IoT environment. The protocol treats AI model parameters as responses and selects challenges capable of generating these responses. Since the responses generated by the challenges can be considered as randomly generated, transmitting the challenges does not disclose the response information, thus avoiding the risk of parameter hacking. Additionally, the protocol utilizes machine-learning modeling techniques and lightweight encryption technologies to reduce the storage costs for identity information and the computational overhead of encryption operations. Through a security analysis of the protocol, it demonstrates that even under ideal attack conditions, the proposed protocol can resist various attacks. By using formal verification with the ProVerif tool, it confirms the security of the protocol flow and the effective protection of private information.

Sequential matching, which aims to predict the item a user will next interact with in the sequential context of the user's historical behaviors, is widely adopted in recommender systems. Existing works mainly characterize the sequential context as the dependencies of user interactions, which is less effective for online travel recommendation where users' behaviors are highly correlated with theirstages in the travel life cycle. Specifically, users on an online travel platform (OTP) usually go through different stages (e.g., exploring a destination, planning an itinerary), and make several correlated interactions (e.g., booking a flight, reserving a hotel, renting a car) at each stage. In this paper, we propose to capture the deep sequential context by modeling the evolving of user stages, and develop a novel stage-aware deep sequential matching network (SASNet) that incorporates inter-stage and intra-stage dependencies over stage-augmented interaction sequence for more accurate and interpretable recommendation. Extensive experiments on real-world datasets validate the superiority of our model for both online travel recommendation and general next-item recommendation. Our model has been successfully deployed at Fliggy, one of the most popular OTPs in China, and shows good performance in serving online traffic.

Recommender Systems (RS), as an efficient tool to discover users' interested items from a very large corpus, has attracted more and more attention from academia and industry. As the initial stage of RS, large-scale matching is fundamental yet challenging. A typical recipe is to learn user and item representations with a two-tower architecture and then calculate the similarity score between both representation vectors, which however still struggles in how to properly deal with negative samples. In this paper, we find that the common practice that randomly sampling negative samples from the entire space and treating them equally is not an optimal choice, since the negative samples from different sub-spaces at different stages have different importance to a matching model. To address this issue, we propose a novel method named Unbiased Model-Agnostic Matching Approach (UMA2). It consists of two basic modules including 1) General Matching Model (GMM), which is model-agnostic and can be implemented as any embedding-based two-tower models; and 2) Negative Samples Debias Network (NSDN), which discriminates negative samples by borrowing the idea of Inverse Propensity Weighting (IPW) and re-weighs the loss in GMM. UMA$^2$ seamlessly integrates these two modules in an end-to-end multi-task learning framework. Extensive experiments on both real-world offline dataset and online A/B test demonstrate its superiority over state-of-the-art methods.

Online-expansion, online maintenance and fault tolerance are vital functions of rail signal monitoring system, which can guarantee the system always in working status and satisfy the sustainable development of rail transit. Due to the limitations of traditional centralized system structure, it is necessary to optimize the design of rail station signal monitoring system to make the system possess these functions. In this research, the novel concept of Role on Autonomous Decentralized System (RoADS) is adopted to optimize the design of rail signal monitoring system. On the one hand, RoADS has the characteristic of role dynamic conversion which is essential to realize the special function of online-expansion, online maintenance and fault tolerance. On the other hand, RoADS also can support the basic function of data collection. In order to verify the validity of the proposed method, a case study has been performed on ZPW-2000 track circuit monitoring system which is the subsystem of rail signal system mainly responsible to monitor the track circuit. The results clearly indicate that the optimized system based on RoADS can not only realize the function of data collection, but also possess the function of online-expansion, online maintenance and fault tolerance. Consequently, the proposed method based on RoADS can improve the safety level of rail transit and reduce the cost operation which is of great significance for maintaining the sustainable urban development of rail transit.

This paper presents Zidian, a middleware for key-value (KV) stores to speed up SQL query evaluation over NoSQL. As opposed to common practice that takes a tuple id or primary key as key and the entire tuple as value, Zidian proposes a block-as-a-value model BaaV. BaaV represents a relation as keyed blocks ( k, B ), where k is a key of a block (a set) B of partial tuples. We extend relational algebra to BaaV. We show that under BaaV, Zidian substantially reduces data access and communication cost. We provide characterizations (sufficient and necessary conditions) for (a) result-preserving queries, i.e., queries covered by available BaaV stores, (b) scan-free queries, i.e., queries that can be evaluated without scanning any table, and (c) bounded queries, i.e., queries that can be answered by accessing a bounded amount of data. We show that in parallel processing, Zidian guarantees (a) no scans for scan-free queries, (b) bounded communication cost for bounded queries; and (c) parallel scalability, i.e., speed up when adding processors. Moreover, Zidian can be plugged into existing SQL-over-NoSQL systems and retains horizontal scalability. Using benchmark and real-life data, we empirically verify that Zidian improves existing SQL-over-NoSQL systems by 2 orders of magnitude on average.

We demonstrate BEAS, a prototype system for querying relations with bounded resources. BEAS advocates an unconventional query evaluation paradigm under an access schema A, which is a combination of cardinality constraints and associated indices. Given an SQL query Q and a dataset D, BEAS computes Q(D) by accessing a bounded fraction DQ of D, such that Q(DQ) = Q(D) and DQ is determined by A and Q only, no matter how big D grows. It identifies DQ by reasoning about the cardinality constraints of A, and fetches DQ using the indices of A. We demonstrate the feasibility of bounded evaluation by walking through each functional component of BEAS. As a proof of concept, we demonstrate how BEAS conducts CDR analyses in telecommunication industry, compared with commercial database systems.