Weekly Update for September 24

class: center, middle, inverse, title-slide

# Weekly Update for September 24
### 2018/09/24

---

class: inverse
# About Slides

---
class: primary 
# New CSAFE slide template

We're now using [`xaringan`](https://github.com/yihui/xaringan)

What's changed:

- New person slide:

````
---
class: inverse
# Your Name
````

---
class: secondary

- New content slide with title:

````
---
class: primary
# Title of slide

Slide content
````

- New content slide without title:

````
---
class: secondary

Slide content with no title on slide
````

---
class: inverse
# Sample User

---
class: primary
# Sample Slide

- Sample Table. Sam Tyner talked about the three must-haves of your summary:

---
class: secondary

- If you are going to add an image, **create a directory** with your name within "images" folder. For example, "images/**guillermo**/sample_image.JPG"

---
class: inverse
# Soyoung

---
class: primary
# Fall semester plans: 
 - Shoes
    + Wrap up shoe analysis using edges and SURF 
    + Compare the performance with other methods such as phase only correlations, Fourier-Mellin transformation correlation
    + Do analysis with CSAFE shoe data
    + Submit papers regarding shoe analysis soon!
   
- Research on "uncertainty pyramid" with Steve and Hari at NIST   
 - Glass analysis with Sam
    + Exploratory analysis 
    + Various covariance estimations
    + Submit papers!
   
- Working on book chapters of glass and shoes funded by ROpenSci Fellowship

---
class: primary
# Analysis result with CSAFE shoe data 
 - All nike shoes with size of 8.5 and 10.5 at time 4 (6 months)
- KM (716) : Between replicates
- KNM (599) : Between shoe IDs
- KNM example between ID 1 and ID2 of size 10.5 at time 4
 <img src="images/soyoung/edge_ex.PNG" width="60%">

---
class: primary
# Analysis result with CSAFE shoe data 
 - All nike shoes with size of 8.5 and 10.5 at time 4 (6 months)
- KM (716) : Between replicates
- KNM (599) : Between shoe IDs
- Edge matching with three circles
 <img src="images/soyoung/nike_6plots.png" width="80%">

---
class: inverse
# Nate

---
class: primary
# Groove ID
- Version 2 of changepoint algorithm running (hopefully bug free)
    - Preliminary results seem a little too variable
- Running algorithm on entire Hamby44 set today
- Next step is to gather all of the code in an R package and thoroughly document everything.

---
class: primary

# SLR
- Looking into "mutual information" between the distribution of the data given a score vs the unconditional distribution of the data to measure information loss 
    - Can potentially estimate this numerically
    
- Putting a prior on `\(H_0\)` vs `\(H_A\)` results in an interesting(?) equality
    - `\(\log(LR) - \log(SLR) = \frac{1}{p(H_A | X,S)} \left[KL(p(M|X,S) || p(M|S)) - \log(\frac{p(H_0|X,S)}{p(H_0|S)}) \right]\)`
    - Is it possible to compute a reasonable upper bound for this, specific to the score distribution of interest? 
    - What does it mean that the left side is independent of prior, but right side is not?

---
class: inverse
# Susan

---
class: primary
# Bullets

- Hamby 44 analysis - compare Hamby 36 to Hamby 44
    - ~~Hopefully will be finished later today~~     
    Finished last Monday. Results are a little weird, but not too bad.
- Resampling with `bulletsamplr` package:     
Use time-series bootstrap methods that ensure continuity (threshold bootstrap) to build a bullet signature using pieces of previously generated signatures
    - Did this week: Got a database of cycles populated with Hamby 36 and Hamby 44 signatures
    - Short term goal: Get the methods working
    - Long term goal: Calculate truly "random" match score statistics to get null distributions