Dynamic Searchable Symmetric Encryption (DSSE) allows clients to update data and search keywords securely over symmetrically encrypted data on an honest but curious server. Conjunctive DSSE, an attractive type of DSSE with expressive search, enables clients to find data containing multiple keywords simultaneously. However, recently proposed efficient conjunctive DSSE schemes, such as ODXT (in NDSS’21) and SDSSE-CQ (in PETS’25), all rely on cross-tag techniques and suffer from either forward privacy or volume-privacy leakages arising from conjunctive keywords, making them vulnerable to injection or leakage-abuse attacks. In this work, we analyze the aforementioned works in depth and design a new conjunctive DSSE scheme named FDXT. For any search query with multiple keywords, FDXT guarantees the forward privacy of all queried keywords. In contrast, ODXT only maintains the forward privacy of the single and first queried keyword. FDXT also avoids volume leakage compared with SDSSE-CQ. Finally, we compared FDXT with ODXT and SDSSE-CQ in terms of performance on the Crime, Wikipedia, and Enron datasets. The experimental results show that FDXT exhibits good performance, which is comparable to ODXT and significantly better than SDSSE-CQ.

Group signatures are active cryptographic topics where group members are granted right to sign messages anonymously on behalf of their group. However, in practical applications, such rights are not permanent in most cases and are usually limited to some time periods. This means that the signing right of each group member needs to be associated with time periods such that it can be automatically changed with the latter. Among the known approaches, verifier local revocation (VLR) seems to be the feasible one to implement the above functionality, but it will cause an inefficient verification process when the group size is large. In this paper, we describe a group signature scheme with time-bound keys, based on the hardness of lattice assumption, which implements the limitation of the signing right to any time period by constructing a lattice-based redactable signature scheme. Our scheme still adds VLR mechanism for some members who need to revoke prematurely, but the time-bound keys function ensures such members are only a small fraction that do not incur excessive cost for revocation check. We give implementation for our scheme under 93-bit and 207-bit security respectively to demonstrate the practicability – all costs are independent of the group size and achieve a relatively efficient level.

We address the general object tracking with arbitrary shape using rangefinders, which is a key module for detecting surrounding traffic and infrastructure for an autonomous driving vehicle. An Expectation-Maximization (EM) algorithm with locally matching is proposed for motion estimation between two consecutive range images. The complexity of the algorithm is O(N) with N the numbers of scan points. Quantitative performance evaluation of the algorithm using a benchmarking vehicular data set. Results of road tests show the effectiveness and efficiency of the implemented system.

In this paper, we propose a novel approach of simultaneous localized feature selection and model detection for unsupervised learning. In our approach, local feature saliency, together with other parameters of Gaussian mixtures, are estimated by Bayesian variational learning. Experiments performed on both synthetic and real-world data sets demonstrate that our approach is superior over both global feature selection and subspace clustering methods.

Typical unsupervised feature selection algorithms select a common feature subset for all the clusters. Consequently, clusters embedded in different feature subspaces are not discovered. In this paper, we propose a novel approach of simultaneous localized feature selection and model detection for unsupervised learning. In our approach, local feature saliency, together with other parameters of Gaussian mixtures, are estimated by Bayesian variational learning. Experiments performed on real-world datasets illustrate that our approach is superior over both global feature selection and subspace clustering methods.

In semi-supervised clustering, domain knowledge can be converted to constraints and used to guide the clustering. In this paper we propose a feature selection algorithm for semi-supervised clustering. In our method, features are conditionally independent. Feature saliency is first computed in unsupervised clustering using the expectation maximization model. Then, it is refined in the tuning step to minimize the feature-wise constraint violation measure, calculated based on the Jensen-Shannon divergence. Experimental results show that a small amount of supervision can improve the performance of clustering and feature selection.

In clustering, global feature selection algorithms attempt to select a common feature subset that is relevant for all clusters. Consequently, they are not able to identify individual clusters that exist in different feature subspaces. In this paper, we propose a localized feature selection algorithm for clustering. The proposed algorithm computes adjusted and normalized scatter separability for individual clusters. A sequential backward search is then applied to find the optimal (maybe local) feature subsets for each cluster. Experiment results on both synthetic data clustering and content-based image grouping show the need for feature selection in clustering and the benefits of selecting features locally.

Top-down induction of decision trees is a simple and powerful method of pattern classification. In a decision tree, each node partitions the available patterns into two or more sets. New nodes are created to handle each of the resulting partitions and the process continues. A node is considered terminal if it satisfies some stopping criteria (for example, purity, i.e., all patterns at the node are from a single class). Decision trees may be univariate, linear multivariate, or nonlinear multivariate depending on whether a single attribute, a linear function of all the attributes, or a nonlinear function of all the attributes is used for the partitioning at each node of the decision tree. Though nonlinear multivariate decision trees are the most powerful, they are more susceptible to the risks of overfitting. In this paper, we propose to perform model selection at each decision node to build omnivariate decision trees. The model selection is done using a novel classifiability measure that captures the possible sources of misclassification with relative ease and is able to accurately reflect the complexity of the subproblem at each node. The proposed approach is fast and does not suffer from as high a computational burden as that incurred by typical model selection algorithms. Empirical results over 26 data sets indicate that our approach is faster and achieves better classification accuracy compared to statistical model select algorithms.

Decision trees represent a simple and powerful method of induction from labeled examples. Univariate decision trees consider the value of a single attribute at each node, leading to the splits that are parallel to the axes. In linear multivariate decision trees, all the attributes are used and the partition at each node is based on a linear discriminate (a hyperplane). Nonlinear multivariate decision trees are able to divide the input space arbitrarily based on higher order parameterizations of the discriminate, though one should be aware of the increase of the complexity and the decrease in the number of examples available as moves further down the tree. In omnivariate decision trees, the decision node may be univariate, linear, or nonlinear. Such architecture frees the designer from choosing the appropriate tree type for a given problem. In this paper, we propose to do the model selection at each decision node based on a novel classifiability measure when building omnivariate decision trees. The classifiability measure captures the possible sources of misclassification with relative ease and is able to accurately reflect the complexity of subproblems at each node. The proposed approach does not require the time consuming statistic tests at each node and therefore does not suffer from as high computational burden as typical model selection algorithm. Our simulation results over several data sets indicate that our approach can achieve at least as good classification accuracy as statistical tests based model select algorithms, but in much faster speed.