- Neural Networks and Applications
- EEG and Brain-Computer Interfaces
- Neural dynamics and brain function
- Mining Techniques and Economics
- Blind Source Separation Techniques
- Color Science and Applications
- Spatial and Cultural Studies
- Reinforcement Learning in Robotics
- Mineral Processing and Grinding
- Complex Network Analysis Techniques
- Fractal and DNA sequence analysis
- Robot Manipulation and Learning
- Belt Conveyor Systems Engineering
- Insect-Plant Interactions and Control
- Visual perception and processing mechanisms
- QR Code Applications and Technologies
- Advanced Optical Imaging Technologies
- Fossil Insects in Amber
- Urban Planning and Landscape Design
- Orthoptera Research and Taxonomy
- Model Reduction and Neural Networks
- Network Security and Intrusion Detection
- Human-Automation Interaction and Safety
- Advanced Graph Neural Networks
- Aesthetic Perception and Analysis
University of Nebraska–Lincoln
2018-2021
McGill University
2016-2021
University of the West of England
2021
Georgia Institute of Technology
2017-2020
Abstract Laser powder bed fusion (LPBF) is an additive manufacturing (AM) process that promises to herald a new age in by removing many of the design and material-related constraints traditional subtractive formative processes. However, level severity defects observed parts produced current class LPBF systems will not be tolerated safety-critical applications. Hence, there need introduce information-rich monitoring assess part integrity simultaneously with fabrication so opportune corrective...
We are working on a scalable, interactive visualization system, called Carina, for people to explore million-node graphs. By using latest web browser technologies, Carina offers fast graph rendering via WebGL, and works across desktop (via Electron) mobile platforms. Different from most existing tools, does not store the full in RAM, enabling it work with graphs up 69M edges. improve open-source offer researchers practitioners new, scalable way visualize large datasets.
We contribute a study benchmarking the performance of multiple motion-based learning from demonstration approaches. Given number and diversity existing methods, it is critical that comprehensive empirical studies be performed comparing relative strengths these techniques. In particular, we evaluate four approaches based on properties an end user may desire for real-world tasks. To perform this evaluation, collected data nine participants, across manipulation The resulting demonstrations were...
In recent years, multivariate pattern analysis (MVPA) has been hugely beneficial for cognitive neuroscience by making new experiment designs possible and increasing the inferential power of functional magnetic resonance imaging (fMRI), electroencephalography (EEG), other neuroimaging methodologies. a similar time frame, “deep learning” (a term use artificial neural networks with convolutional, recurrent, or similarly sophisticated architectures) produced parallel revolution in field machine...
It remains unclear how the visual system is able to extract affective content from complex scenes even with extremely brief (< 100 millisecond) exposures. One possibility, suggested by findings in machine vision, that low-level features such as unlocalized, two-dimensional (2-D) Fourier spectra can be diagnostic of scene content. To determine whether image amplitude carries any information about quality scenes, we first validated existence category differences through a support vector (SVM)...
Virtual reality (VR) and personal head-mounted displays (HMDs) can be a viable tool for the presentation of scientifically accurate valid demonstrative data in courtroom. However, capabilities limitations technology need to fully characterized. The current pilot study evaluated visual acuity contrast sensitivity using two commercially available HMDs (Oculus Rift HTC Vive Pro). Preliminary findings indicated that experienced VR may less than what is real-world scenarios. provides quantitative...
Many recent developments in machine learning have come from the field of "deep learning," or use advanced neural network architectures and techniques. While these methods produced state-of-the-art results dominated research focus many fields, such as image classification natural language processing, they not gained much ground over standard multivariate pattern analysis (MVPA) techniques electroencephalography (EEG) other human neuroscience datasets. The high dimensionality large amounts...
This paper describes a system level approach to achieve flash memory that would consume very little current (less than 1μA) in standby mode and wake up fast (~1μs) for random-access read operation. The mainly focuses how analog circuits other components can be partitioned these specifications. In addition, design considerations various have also been illustrated.
Abstract In recent years, multivariate pattern analysis (MVPA) has been hugely beneficial for cognitive neuroscience by making new experiment designs possible and increasing the inferential power of functional magnetic resonance imaging (fMRI), electroencephalography (EEG), other neuroimaging methodologies. a similar time frame, “deep learning” (a term use artificial neural networks with convolutional, recurrent, or similarly sophisticated architectures) produced parallel revolution in field...
Abstract Malicious software, known as malware, is a perpetual game of cat and mouse between malicious software developers security professionals. Recent years have seen many high profile cyber attacks, including the WannaCry NotPetya ransomware attacks that resulted in major financial damages to businesses institutions. Understanding characteristics such how malware can propagate interact systems networks key for mitigating these threats containing infection avoid further damage. In this...
Abstract Many signals, particularly of biological origin, suffer from a signal-to-noise ratio sufficiently low that it can be difficult to classify individual examples reliably, even with relatively sophisticated machine-learning techniques such as deep learning. In some cases, the noise high enough is achieve convergence during training. We considered this problem for one data type often suffers difficulties, namely electroencephalography (EEG) cognitive neuroscience studies in humans. One...
We assayed the contributions of image Fourier amplitude spectra (AS) and color in two experiments focusing on rapid categorization affective versus neutral natural scenes. Focusing initial feed-forward sweep activation through visual hierarchy, we used briefly flashed (~33 ms) scenes that were immediately backward masked with textures. Previous studies hint low-level AS information might guide detection some categories (e.g., human faces). In Experiment 1, a method developed by Gaspar...
The ability to accurately depict the scene of an incident a jury, who will never be able experience it first-hand, is key element presenting trial demonstratives. In many cases, where visual perception and situational awareness are in understanding nature incident, traditional modes presentation, such as two-dimensional photographs, may insufficient require more sophisticated methodology. Specifically, stereoscopic head-mounted displays (HMDs) virtual reality (VR) programming continues...