Projects

Abstract: Graph-based clustering partitions data by similarity graph and performs well on many tasks. When data has noise and outliers, it’s hard to make a good similarity graph, which determines clustering results. This article proposes…

Abstract: Time-driven learning updates prediction model parameters as new data arrives in this work. The direct heuristic dynamic programming (dHDP) algorithm has been shown to solve several complex learning control problems. As system states change,…

Abstract: Online autonomous learning algorithms mine data streams with minimal computational complexity. Due to their structural simplicity, parsimonious learning machines (PALMs) are suitable for such applications. These parsimonious algorithms use thresholds to add or remove…

Abstract: Machine learning’s multi-scale decision system (MDS) describes hierarchical data well. MDS knowledge discovery involves OSC selection and attribute reduction. Searching for all OSCs may cause a combinatorial explosion, and current methods are time-consuming. This…

Abstract: This article examines neuroscience, network science, and dynamic systems research on brain-inspired intelligence. Instead of using artificial computing networks, reconstructing cortical networks with dynamic activities that nourish brain functions can replicate brain intelligence. The…

Abstract: Resting-state functional magnetic resonance imaging (rs-fMRI) functional connectivity (FC) networks can diagnose Alzheimer’s disease and its prodromal stage, mild cognitive impairment (MCI). FC is estimated as a temporal correlation of regional mean rs-fMRI signals…

Abstract: Noninvasive structural magnetic resonance imaging (MRI) can accurately predict neuropsychological test scores, which helps diagnose dementia (e.g., Alzheimer’s disease (AD)) and predict its progression. Existing machine/deep learning approaches preselect dementia-sensitive brain locations for MRI…

Abstract: In our data world, many untrusted controllers collect personal data. Giving the individual data agency protects her privacy and informational self-determination. Local anonymization allows each person to anonymize her data before giving them to…

Abstract: Kernel-based extreme learning machine (KELM), a natural extension of ELM to kernel learning, excels in regression and classification problems. Due to its random projection mechanism and lack of hidden node requirements, KELM generalizes better…

Abstract: With the maturity and popularity of Internet of Things (IoT), Social Internet of Things (SIoT) has been proposed to support novel IoT applications and networking services more effectively and efficiently. Many SIoT works design…

Abstract: Textual information helps model item concepts. Many models ignore conceptual “correlation sparsity” by not using concepts’ inferential property to capture word meanings. This paper distinguishes between word modeling and concept modeling and proposes an…

Abstract: This paper addresses static 3D cloth draping on virtual human bodies. A two-stream deep network model that extracts features from body and garment shapes drapes a template cloth on virtual 3D bodies. Our network…