Neurosymbolic (NeSy) AI combines neural architectures and symbolic reasoning to improve accuracy, interpretability, and generalization. While logic inference on top of subsymbolic modules has been shown to effectively guarantee these properties, this often comes at the cost of reduced scalability, which can severely limit the usability of NeSy models. This paper introduces DeepProofLog (DPrL), a novel NeSy system based on stochastic logic programs, which addresses the scalability limitations of previous methods. DPrL parameterizes all derivation steps with neural networks, allowing efficient neural guidance over the proving system. Additionally, we establish a formal mapping between the resolution process of our deep stochastic logic programs and Markov Decision Processes, enabling the application of dynamic programming and reinforcement learning techniques for efficient inference and learning. This theoretical connection improves scalability for complex proof spaces and large knowledge bases. Our experiments on standard NeSy benchmarks and knowledge graph reasoning tasks demonstrate that DPrL outperforms existing state-of-the-art NeSy systems, advancing scalability to larger and more complex settings than previously possible.

There is increasing interest in employing large language models (LLMs) as cognitive models. For such purposes, it is central to understand which properties of human cognition are well-modeled by LLMs, and which are not. In this work, we study the biases of LLMs in relation to those known in children when solving arithmetic word problems. Surveying the learning science literature, we posit that the problem-solving process can be split into three distinct steps: text comprehension, solution planning and solution execution. We construct tests for each one in order to understand whether current LLMs display the same cognitive biases as children in these steps. We generate a novel set of word problems for each of these tests, using a neuro-symbolic approach that enables fine-grained control over the problem features. We find evidence that LLMs, with and without instruction-tuning, exhibit human-like biases in both the text-comprehension and the solution-planning steps of the solving process, but not in the final step, in which the arithmetic expressions are executed to obtain the answer.

Abstract In order to improve transmission performance, opportunistic social networks are widely used in 5G mobile network communication for network congestion. However, transferring large amounts of data in an instant will lead to data redundancy problems, which will affect the transmission quality. The authors propose a method called user‐optimized data transmission scheduling based on edge community service in opportunistic social networks (ECSUO). This method constructs the edge community service model by introducing mobile edge computing into the opportunistic social network. Then several scheduling factors are set to coordinate the dynamic priority assessment. The simulation results show that ECSUO performs better than other classical opportunistic complex network algorithms in a variety of transmission quality evaluation indexes, and this method also has good applicability when applied to different transmission models.

Abstract With the advent of the 5G era, people's demand for information and bandwidth is increasing. The signal source will mainly be high‐frequency radio waves with high bandwidth. However, the coverage of high‐frequency radio waves is very limited, which leads to the emergence of micro base stations. Opportunistic social networks are just suitable for this new transmission method, it is easy for users to simultaneously request content with high popularity in a short time, which will cause a heavy load on the micro base station and a high transmission delay. This paper establishes the community mechanism and proposes a user cache management and cooperative transmission mechanism (CMCT) based on edge community computing by using the predictability of users' requests for popular content. The free user equipment space is used to manage cache distribution through the edge community computing to perform cooperative transmission, thereby reducing the content delivery delay and the load of the micro base station. Simulation experiments show that the CMCT algorithm can reduce end‐to‐end delay and overhead, and improve data transmission efficiency.

Abstract Wireless sensor networks have been used widely in environmental tracking and monitoring. Nevertheless, WSNs are facing a number of challenges, such as unreasonable cluster head selection and energy‐hole problems while nodes have unbalanced energy consumption. Thus, the paper aims to propose an energy‐efficient non‐uniform clustering routing protocol to enhance nodes energy efficiency and balance the energy consumption in WSNs. In addition, a non‐uniform clustering network partition is introduce to reduce the probability of energy hole occurrences and optimize cluster heads dynamical selection method to suggest an improved shuffled frog leaping algorithm. Ultimately, the simulation experiment has demonstrated a 20% gain of energy efficiency to extend network lifetime by the proposed protocol and enhanced algorithm.

This paper describes an improved synchronization method for MC-CDMA (multi-carrier code division multiple access) systems. The method takes advantage of the properties of OFDM signals and spreading signals without any additional pilots or training sequences. It mainly discusses symbol timing recovery from the time domain and carrier frequency offset from the frequency domain.

Multicarrier code division multiple access (MC-CDMA) technology, which combines advantages of both OFDM and CDMA, has superior performance when transmitting high rate data in wireless fading channels. However, synchronization performance plays a key role on MC-CDMA systems. Based on the analysis of present synchronization technologies, a new synchronization algorithm is proposed. Theoretical analysis and computer simulation show that this algorithm has superior synchronization performance and less system complexity.