Video-based multimodal depression estimation is a challenging task that relies on comprehensive video understanding and analysis to predict depression severity. This process involves capturing discriminative information from multimodal features, including auditory, textual, and visual signals in the video data. Existing foundation models primarily focus on fusing multimodal features but often overlook two essential aspects: (1) the alignment of intra- and inter-modality and (2) preserving modality-specific semantic information while eliminating redundancy . Hence, these issues inevitably result in semantic misalignment between modalities, leading the model to learn redundant or even conflicting representations. To tackle these challenges, we propose a Contrastive Modality Alignment and Fusion (CMAF) framework. Specifically, each modality is divided into multiple segments, and a dynamic normalization strategy is applied to align the temporal context within each modality. Simultaneously, contrastive learning is utilized to achieve semantic alignment across different modalities. Together, these components form the Contextual Modality Alignment (CMA) module. Furthermore, a Cross-Attention Contrastive Fusion (CCF) module is presented, where cross-attention is used to achieve complementarity and enhancement between modalities, while contrastive learning further promotes modal consistency and preserves modality-specific information without redundancy. Therefore, the fused representation can integrate the features of each modality without losing the specific information of the unimodal. Extensive experiments on public datasets, including CMDC, DAIC-WOZ, and E-DAIC, demonstrate that CMAF outperforms advanced depression estimation methods and highlights the essence of effective alignment and fusion.

Micro-expressions are typically regarded as unconscious manifestations of a person's genuine emotions. However, their short duration and subtle signals pose significant challenges for downstream recognition. We propose a multi-task learning framework named the Adaptive Motion Magnification and Sparse Mamba (AMMSM) to address this. This framework aims to enhance the accurate capture of micro-expressions through self-supervised subtle motion magnification, while the sparse spatial selection Mamba architecture combines sparse activation with the advanced Visual Mamba model to model key motion regions and their valuable representations more effectively. Additionally, we employ evolutionary search to optimize the magnification factor and the sparsity ratios of spatial selection, followed by fine-tuning to improve performance further. Extensive experiments on two standard datasets demonstrate that the proposed AMMSM achieves state-of-the-art (SOTA) accuracy and robustness.

This article presents a rolling optimization algorithm (ROA) for minimizing total vehicle delay for real-time traffic signal control in urban road networks, where the vehicle delay minimization problem is formulated as a quadratically constrained quadratic programming problem, which is NP-hard. The programming problem is relaxed and resolved using the ROA in a distributed manner. In particular, we introduce network partition to a large-scale urban road network and then present a regional ROA to eliminate the low efficiency caused by the large number of decision variables. Numerical experiments are performed on one of Beijing’s district road networks to validate the efficiency of the proposed method in benchmark against several typical techniques.

Mathematical reasoning aims to infer satisfiable solutions based on the given mathematics questions.Previous natural language processing researches have proven the effectiveness of sequence-to-sequence (Seq2Seq) or related variants on mathematics solving.However, few works have been able to explore structural or syntactic information hidden in expressions (e.g., precedence and associativity).This dissertation set out to investigate the usefulness of such untapped information for neural architectures.Firstly, mathematical questions are represented in the format of graphs within syntax analysis.The structured nature of graphs allows them to represent relations of variables or operators while preserving the semantics of the expressions.Having transformed to the new representations, we proposed a graph-to-sequence neural network GraphMR, which can effectively learn the hierarchical information of graphs inputs to solve mathematics and speculate answers.A complete experimental scenario with four classes of mathematical tasks and three Seq2Seq baselines is built to conduct a comprehensive analysis, and results show that GraphMR outperforms others in hidden information learning and mathematics resolving.

Satisfiability Modulo Theories (SMT) solvers have been widely applied in automated software analysis to reason about the queries that encode the essence of program semantics, relieving the heavy burden of manual analysis. Many SMT solving techniques rely on solving Boolean satisfiability problem (SAT), which is an NP-complete problem, so they use heuristic search strategies to seek possible solutions, especially when no known theorem can efficiently reduce the problem. An emerging challenge, named Mixed-Bitwise-Arithmetic (MBA) obfuscation, impedes SMT solving by constructing identity equations with both bitwise operations (and, or, negate) and arithmetic computation (add, minus, multiply). Common math theorems for bitwise or arithmetic computation are inapplicable to simplifying MBA equations, leading to performance bottlenecks in SMT solving.

Since the State Council of China issued the “Outline of Action to Promote the Development of Big Data” in 2015, which proposed a top-level design for big data development from a long-term perspective, the related industry had ushered in a rapid way. At present, under the background of accelerating digital transformation and post COVID-19, the social and economic benefits of data assets, especially the value of financial data turn out to be more prominent. In this paper, we analyze four exploration aspects of commercial banks around big data application: constructing enterprise-level big data platforms, strengthening the capability of risk control, innovating financial products and services, optimizing business processes. Meanwhile, giving the typical cases for all of these practices, we illustrate how the bank's decision-making transformed from “experience-dependent” to “data-based” that enable banks to improve business performance, assess credit risks and allocate resources scientifically. Generally, we hold the view that big data technology has become an important means for commercial banks to meet the emerging financial demands, respond to the competition of the fintechs and the high-level opening-up trend. At last, this paper suggests banks enhance the core competitiveness through big data deep application with facing challenges such as insufficient implementation of big data strategy, lack of data governance system, imperfect organizational structure as well as scarcity of data mining talents.

The emerging of machine learning has massively promoted the abilities of computational sustainability in natural resource management and allocation. Many Internet giants such as Google, Amazon, and Microsoft now provide Machine Learning as a Service (MLaaS) to meet the increasing demand for machine learning services. However, the prediction results of training data and testing data with the same machine learning model in MLaaS have remarkable differences, and thus the attackers can leverage machine learning techniques to launch the so-called membership inference attacks, i.e., to infer whether a record is in the training data or not. In this paper, we propose MIASec that can guarantee the data indistinguishability of the training data and thereby has the ability to defend against membership inference attacks in MLaaS. The key idea of MIASec is to narrow the dynamic ranges of vital features in the training data, such that the training data, the testing data, and even the synthetic data have almost semblable prediction results by the same machine learning model. With elaborated design on modifying the values of vital features in the training data, MIASec can thus reduce the differences between the model's outcomes of training data and testing data, thereby protecting the training data in effect while keeping the model's accuracy stable. We empirically evaluate MIASec on machine learning models trained by off-line neural networks and on-line MLaaS. Using realistic data and classification tasks, our experiment results show that MIASec can defend the membership inference attacks effectively. In particular, MIASec can reduce the precision and recall of attacks respectively by 11.7 and 15.4 percent in average, and by 18.6 and 21.8 percent at best.