annapoliswu · pvhau169 · Aug 26, 2021 · Aug 26, 2021 · Aug 26, 2021 · Aug 27, 2021
diff --git a/.DS_Store b/.DS_Store
diff --git a/.gitignore b/.gitignore
@@ -0,0 +1,16 @@
+
+public/.DS_Store
+.DS_Store
+public/.DS_Store
+.DS_Store
+.DS_Store
+content/.DS_Store
+public/.DS_Store
+public/images/.DS_Store
+static/.DS_Store
+*.log
+public/.DS_Store
+public/images/.DS_Store
+static/.DS_Store
+public/.DS_Store
+static/.DS_Store
diff --git a/README.md b/README.md
@@ -1,4 +1,4 @@
-# CNI Engineering Lab Website
+# CNIEL Website
 
 The following website uses the Hugo Framework: (<https://gohugo.io/documentation/>). 
 
@@ -26,7 +26,7 @@ There are four main folders that are used.
   - All page-specfic css styles are in the their respective layout pages.
   - /_default folder
     - index.html is where you can edit the front page
-    - /partials include code blocks that can be inserted into html with {{partial "partialName" .}}
+    - /partials includes code blocks that can be inserted into html with {{partial "partialName" .}}
     - /_markup/render-image.html is used to ensure that all images in the content markup files are rendered correctly in html
   - /singles
     - Specifies the layouts for special one-off pages that need special layouts
@@ -43,10 +43,10 @@ Adding pages is as simple as creating a new file and filling in the information.
 - 'hugo new research/researchName.md'
 - Fill in frontmatter
 - Body of text and images go in the md body
-
+- Pages are sorted in lists by manually changed publish/edit date (this can be used to sort members and research pages)
 ## Miscellaneous Notes
 
-- Filling in the 'externalLink' field in research will specify it as an external research page
+- Filling in an the 'externalLink' field in research will specify it as an external research page
   - This page will not appear in the navbar
   - You only need to fill in title and shortDescription for the list page display
 - Any links on the members page will be automatically converted into link-buttons

diff --git a/archetypes/default.md b/archetypes/default.md
diff --git a/archetypes/members.md b/archetypes/members.md
diff --git a/archetypes/research.md b/archetypes/research.md
diff --git a/config.toml b/config.toml
@@ -1,6 +1,6 @@
-baseURL = "https://annapoliswu.github.io/CNI_Lab/"
+baseURL = "https://www.eecis.udel.edu/~ajbrock/"
 languageCode = "en-us"
-title = "CNI Engineering Lab"
+title = "CNIEL - University of Delaware"
 sectionPagesMenu = "main"
 CanonifyURLs = true
 

diff --git a/content/.DS_Store b/content/.DS_Store
diff --git a/content/members/.DS_Store b/content/members/.DS_Store
diff --git a/content/members/AustinBrockmeier.md b/content/members/AustinBrockmeier.md
@@ -1,17 +1,34 @@
 ---
-name: "Austin J. Brockmeier"
-title: "Some Title"
+name: "Austin J. Brockmeier, Ph.D."
+title: "Assistant Professor"
 imageLink: "/images/members/AustinBrockmeier.jpg"
 currentMember: true
 
 date: 2018-04-28T17:07:03-04:00
 draft: false
 ---
 
-I am an assistant professor in the Department of Electrical and Computer Engineering (ECE) and the Department of Computer and Information Sciences (CIS) within the College of Engineering at the University of Delaware (UD), and a resident faculty in UD's Data Science Institute.
+[Curriculum Vitae](/other/ajbrockmeier_CV_12_2021.pdf)
+[Email](mailto:ajbrock@udel.edu)
+
+*What*: **Data Science, Machine Learning, and Signal Processing**   
+These involve the underlying mathematical analysis, design of statistical models, and software implementation of data and signal processing (filtering and neural networks) as well as optimization routines.
+
+* Finding discrepancies in data sets using [divergences](/research/discrepancies/)
+* Selecting representations of data using [informativeness measures](/research/informativeness/)
+
+*Why*: To extract actionable information from complex data, especially to **understand and interface with the brain**
 
-My research interests include data science, machine learning, signal processing, and the underlying mathematics and design of statistical models, optimizations, and algorithms. I have experience interfacing with complex data from the domains of biomedical engineering, neuroscience, and natural language processing.
+* Identify meaningful features (in space or time) of the evoked neural response using [metric learning](/research/neural_decoding/)
+* Find patterns of waveforms in EEGs using [neural signal processing](/research/eegs/)
+* Extract information from text and use machine learning to prioritize large collections of articles using [natural language processing and text mining](/research/pico/)
 
 [Google Scholar](https://scholar.google.com/citations?hl=en&user=g_QoCQQAAAAJ&view_op=list_works&sortby=pubdate)
-[Curriculum Vitae](https://www.ece.udel.edu/wp-content/uploads/2021/03/ajbrockmeier-cv-03-2021.pdf)
-[Email](mailto:ajbrock@udel.edu)
+
+*Where*: **University of Delaware** 
+* Primary appointment: [Department of Electrical and Computer Engineering (ECE)](https://www.ece.udel.edu) 
+* Joint appointment: [Department of Computer and Information Sciences (CIS)](https://www.cis.udel.edu/)
+* Both are within the [College of Engineering](https://engr.udel.edu/)
+* Resident faculty in UD's [Data Science Institute](https://dsi.udel.edu/)
+
+
diff --git a/content/members/BilalRiaz.md b/content/members/BilalRiaz.md
@@ -1,11 +1,11 @@
 ---
-name: "Bilal Riaz"
-title: "M.S."
+name: "Bilal Riaz, M.S."
+title: "Ph.D. student"
 imageLink: 
 currentMember: true
 
-date: 2021-04-28T17:00:22-04:00
+date: 2019-08-28T17:00:22-04:00
 draft: false
 ---
 
-Unsupervised learning as optimizations with positive semidefinite matrices.
+###### Unsupervised learning as optimizations with positive semidefinite matrices. {#bilal}
diff --git a/content/members/CarlosCardenas.md b/content/members/CarlosCardenas.md
@@ -1,11 +1,14 @@
 ---
-name: "Carlos Mendoza Cárdenas"
-title: "M.S."
-imageLink: 
+name: "Carlos H. Mendoza Cárdenas, M.S."
+title: "Ph.D. student"
+imageLink: "/images/members/Carlos.jpeg"
 currentMember: true
 
-date: 2021-04-28T13:12:57-04:00
+date: 2019-01-01T13:12:57-04:00
 draft: false
 ---
+**Research focus**: finding patterns in neural time series through convolutional sparse analysis. 
 
-Works with spotting brain wave patterns using neural networks.
+**Goal**: to discover physiologically meaningful waveforms in multi-day continuous epileptic electrocorticographic (ECoG) recordings that can be used to build interpretable features for seizure prediction. 
+
+**Methods**: interpretable machine learning, clustering and sparse coding for time series, supervised learning for neural data 
diff --git a/content/members/CesarOlivares.md b/content/members/CesarOlivares.md
@@ -1,11 +1,12 @@
 ---
-name: "C. Cesar Claros Olivares"
-title: "M.S."
-imageLink: 
+name: "C. Cesar Claros Olivares, M.S."
+title: "alumnus and admitted Ph.D. student"
+imageLink: "/images/members/Cesar.jpg"
 currentMember: true
 
-date: 2021-04-28T17:00:02-04:00
+date: 2019-05-01T17:00:02-04:00
 draft: false
 ---
 
-Vigilant machine learning.
+###### Error-aware and vigilant machine learning. {#cesar}
+3D convolutional neural networks for brain age prediction from structural measures.  Optimal transport for color and style transfer.
diff --git a/content/members/EdwinSalcedo.md b/content/members/EdwinSalcedo.md
@@ -1,10 +1,10 @@
 ---
-name: "Edwin Salcedo"
-title: "M.B.A"
-imageLink: 
+name: "Edwin Salcedo, M.Sc., M.B.A."
+title: "Visiting Scholar (Summer 2019)"
+imageLink: "/images/members/Edwin.png"
 currentMember: false
 
-date: 2021-04-28T17:36:38-04:00
+date: 2019-07-01T17:36:38-04:00
 draft: false
 ---
 

diff --git a/content/members/EvanCurtin.md b/content/members/EvanCurtin.md
@@ -0,0 +1,11 @@
+---
+name: "Evan Curtin"
+title: "Undergraduate (Summer Scholar 2021)"
+imageLink: "/images/members/EvanCurtin.png"
+currentMember: true
+
+date: 2021-06-01T21:00:00-04:00
+draft: false
+---
+
+###### Non-Negative Matrix Factorization as Dictionary Learning for Audio Separation
diff --git a/content/members/HassanBaker.md b/content/members/HassanBaker.md
@@ -1,11 +1,11 @@
 ---
-name: "Hassan Baker"
-title: "M.S."
-imageLink: 
+name: "Hassan Baker, M.S."
+title: "Ph.D. student"
+imageLink: "/images/members/Hassan.png"
 currentMember: true
 
-date: 2021-04-28T17:00:38-04:00
+date: 2020-02-01T17:00:38-04:00
 draft: false
 ---
 
-Neural decoding and signal processing.
+###### Graph signal processing for functional and structural neuroimaging. {#hassan}
diff --git a/content/members/HauPhan.md b/content/members/HauPhan.md
@@ -1,11 +1,11 @@
 ---
-name: "Hau Phan"
-title: "M.S."
-imageLink: 
+name: "Hau Phan" 
+title: "M.S. and Ph.D. student"
+imageLink: "/images/members/Hau.gif"
 currentMember: true
 
-date: 2021-04-28T17:00:52-04:00
+date: 2021-01-01T17:00:52-04:00
 draft: false
 ---
 
-Reinforcement learning estimation and detection.
+###### Reinforcement learning estimation and detection. {#hau}
diff --git a/content/members/JustinLabombard.md b/content/members/JustinLabombard.md
@@ -0,0 +1,11 @@
+---
+name: "Justin Labombard"
+title: "Undergraduate (Summer Scholar 2021)"
+imageLink: "/images/members/Justin.png"
+currentMember: true
+
+date: 2021-06-15T17:00:22-04:00
+draft: false
+---
+
+###### Dictionary Learning on Epileptic ECoG Waveforms
diff --git a/content/members/NicolasLopez.md b/content/members/NicolasLopez.md
@@ -0,0 +1,11 @@
+---
+name: "Andres Nicolas Lopez, MSc."
+title: "Visiting Scholar (Summer 2021)"
+imageLink: "/images/members/Nicolas.jpg"
+currentMember: false
+
+date: 2021-06-20T17:36:38-04:00
+draft: false
+---
+
+Statistician working on the error modeling for synergistic machine learning. [LinkedIn](https://www.linkedin.com/in/anlopezl/)
diff --git a/content/members/YukselKarahan.md b/content/members/YukselKarahan.md
@@ -1,11 +1,13 @@
 ---
-name: "Yuksel Karahan"
-title: "M.S."
+name: "Yuksel Karahan, M.S."
+title: "Ph.D. student"
 imageLink: 
 currentMember: true
 
-date: 2021-04-28T16:59:19-04:00
+date: 2019-08-28T16:59:19-04:00
 draft: false
 ---
 
-Data science techniques for semisupervised domain transfer.
+**Research interests**: Data science techniques for semi-supervised domain transfer. Using machine learning tools to understand complex data that includes analyzing and optimizing the  algorithms and their implications on data.
+
+**Current project**: Currently working  on Wasserstein-based divergence measures to understand discrepancy between distributions.
diff --git a/content/research/.neural_decoding.md.swp b/content/research/.neural_decoding.md.swp
diff --git a/content/research/discrepancies.md b/content/research/discrepancies.md
@@ -0,0 +1,28 @@
+---
+title: "Interpretable maximal discrepancies metrics for analyzing and improving generative models"
+shortTitle: "Interpretable Divergences"
+shortDescription: "Project funded by the Office of Naval Research"
+
+authors: "Office of Naval Research, Grant # N00014-21-1-2300, Principal Investigator: Austin J. Brockmeier,  4/2021–4/2024."
+
+imageLink: "/images/research/discrepancies.png"
+externalLink: 
+journalLink: 
+pdfLink:  
+codeLink: 
+
+date: 2021-04-14T00:59:28-04:00
+draft: false
+---
+
+## Overview
+Divergence measures quantify the dissimilarity, including the distance, between distributions and are fundamental to hypothesis testing, information theory, and the estimation and criticism of statistical models. Recently, there has been renewed interest in divergences in the context of generative adversarial neural networks (GANs). While a multitude of divergences exist, they vary in their characteristics. Importantly, not all divergences are equally interpretable: a divergence between samples is considered interpretable if it directly answers the question “Which instances best exhibit the discrepancy between the samples?”
+
+![Where do the distributions differ.](/images/research/discrepancies2.png "Example of discrepancies. ")
+
+#### Outcomes
+* **Y. Karahan**, B. Riaz, and **A. J. Brockmeier** "Kernel landmarks: An empirical statistical approach to detect covariate shift", *Workshop on Distribution Shifts, 35th Conference on Neural Information Processing Systems (NeurIPS 2021)* [Paper](https://openreview.net/forum?id=Wu5hMMQ76OE), [Poster](/other/karahan_poster_distshift2021), [5 min Presentation](https://recorder-v3.slideslive.com/?share=56618&s=79a6864d-0cec-41ad-a152-09cceb1c6bf4).
+*  **A. J. Brockmeier**, **C. C. Claros Olivares**, M. S. Emigh, and  L. G. Sanchez Giraldo "Identifying the instances associated with distribution shifts using the max-sliced Bures divergence", *Workshop on Distribution Shifts, 35th Conference on Neural Information Processing Systems (NeurIPS 2021)* [Paper](https://openreview.net/forum?id=y_s0M6OtyH_), [Poster](/other/brockmeier_poster_distshift2021), [5 min Presentation](https://recorder-v3.slideslive.com/?share=56676&s=b55c603d-1e31-4ab2-9cd7-440e5d05d64d).
+
+#### Acknowledgements 
+These research efforts are sponsored by the Department of the Navy, Office of Naval Research under ONR award number N00014-21-1-2300. Gratefully acknowledge the support of Dr. Tory Cobb and ONR321.  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Office of Naval Research. 
diff --git a/content/research/eegs.md b/content/research/eegs.md
@@ -10,18 +10,22 @@ issue: 1
 pages: "43-54"
 year: 2016
 
-imageLink: "/images/research/eegs1.png"
+imageLink: "/images/research/eegs2.png"
 externalLink:
 journalLink: "http://dx.doi.org/10.1109/TBME.2015.2499241"
-pdfLink: "http://cnel.ufl.edu/files/1447193268.pdf"
+pdfLink: "/papers/brockmeier2016_learning_recurrent_waveforms.pdf"
 codeLink: "/code/eegs.zip"
 
-date: 2021-04-28T22:42:02-04:00
+date: 2016-04-28T22:42:02-04:00
 draft: false
 ---
 
 ## Overview
 
 When experts analyze EEGs they look for landmarks in the traces corresponding to established patterns such as oscillatory and phasic events of particular frequency or morphology. Long records motivate automated analysis techniques. Automation techniques often require design choices such as wavelet family or number of bandpass filters. To overcome this, we explore a modeling approach that automatically learns recurrent temporal waveforms within EEG traces. The estimation is based on a multiple-input, single-output linear model with sparsely excited inputs.
 
-![Figure 1](/images/research/eegs1.png "Figure 1")
+![We assume that an EEG signal can be described by a convolutional sparse coding model. The EEG trace is approximated as an additive mixture of component signals each described by a convolution of a sparse source with a waveform. Although this is a linear model (multiple input and single output), to separate the components requires a nonlinear analysis. Additionally, when the waveforms are unknown, this blind source separation problem is even more challenging.](/images/research/eegs1.png "Convolutional sparse coding model for EEG")
+
+We apply the approach to various data sets to better understand the specificity and consistency of the estimated waveforms. In particular, we cluster the waveforms found on different electrodes/channels and subjects. 
+
+![Each cluster of waveforms is described by its centroid, the waveform nearest the centroid, the spectrum of the centroid, and the spatial distribution of the originating electrodes for the cluster waveforms.](/images/research/eegs2.png "Recurrent waveforms")
diff --git a/content/research/graphs.md b/content/research/graphs.md
@@ -0,0 +1,29 @@
+---
+title: "Advancing machine learning for neuroimaging through topology-aware signal
+processing"
+shortTitle: "Graphs and Signal Processing for Neuroimaging"
+shortDescription: "Project funding from a University of Delaware Research Foundation-Strategic Initiative grant"
+
+authors: "University of Delaware Research Foundation–Strategic Initiative,  Principal Investigator: Austin J. Brockmeier, Mentor: Gonzalo Arce,  11/2019–10/2021."
+
+imageLink: "/images/research/brainGraph1.png"
+externalLink: 
+journalLink: 
+pdfLink:  
+codeLink: 
+
+date: 2020-02-01T00:59:28-04:00
+draft: false
+---
+The proposed work plan is to develop machine learning techniques to work directly with graph signal processing techniques in the context of neuroimaging. The goal is to leverage information in the form of the topology of the signal sensors or measurement locations to refine the neural signal representations in order to improve the statistical power of tests for distinguishing differences between conditions or stimuli. The project’s scope includes the formulation, mathematical and statistical analysis, and initial validation of the proposed methodology.
+
+![First 5 harmonics (Fourier transform) of the gray matter graph.](/images/research/brainGraph2.png "Example of harmonics of gray matter. ")
+
+#### Outcomes
+*  NER'21: **Hassan Baker**, A. J. Brockmeier "Local and Sparse Linear Causal Models for fMRI Resting-State Signals" [Full-text (IEEE)](https://ieeexplore.ieee.org/document/9441242/) | [Poster PDF](/other/baker_poster_NER2021.pdf)
+*Machine Learning for Health (ML4H): **Hassan Baker**, A. J. Brockmeier "Exploring latent networks in resting-state fMRI using voxel-to-voxel causal modeling feature selection" [Paper](https://arxiv.org/pdf/2111.07488.pdf), [Poster](/other/baker_poster_ML4H2021.pdf)
+
+
+#### Acknowledgements 
+Research was carried out with the support of the University of Delaware Research Foundation. Hassan Baker acknowledges support from the Gore Engineering
+Graduate Fellowship in Spring 2021. This research was supported in part through the use of Data Science Institute (DSI) computational resources at the University of Delaware, specifically the DARWIN system. The DARWIN computing project at the University of Delaware is supported by the National Science Foundation under Grant No. OAC-1919839. The authors thank the support from the University of Delaware IT Research Computing Group.
diff --git a/content/research/informativeness.md b/content/research/informativeness.md
@@ -16,7 +16,7 @@ journalLink: "http://www.jmlr.org/papers/v18/16-296.html"
 pdfLink: "http://www.jmlr.org/papers/volume18/16-296/16-296.pdf"
 codeLink: "/code/informativeness.zip"
 
-date: 2021-03-17T21:25:29-04:00
+date: 2017-03-17T21:25:29-04:00
 draft: false
 ---