Official CBSA video on Improving Technology for the border:
https://www.cbsa-asfc.gc.ca/multimedia/it-pt/menu-eng.html

This presentation is not associated with CBSA.
All references to CBSA are obtained from Public domain.

• What is AI and NN ?
• Background
  • Video Recognition Systems (VRS)
  • Fully-connected pseudo-inverse neural networks (PINN)
• How NN (AI) was done in the past
  • Success based on intelligent preprocessing
  • Demo: Face Recognition in Video (FRiV)
• How AI (NN) is done now
  • Old FRiV ideas implemented via new Keras functionality
  • Good and bad use of NN: Sequencial data example
• Vision for AI and AI scientist
  • Intelligent multilingual text analysis example

• Artificial Intelligence (AI)
• Neural networks (NN)
• Highly complex, very diverse, densly-connected, recurrent, multi-layered neural networks,
• Plus, leveraging many other Machine Learning and Stats techniques

My former colleagues: Ivakhnenko: library(GMDH), Kussul, Reznik

To built highly intelligent systems

• capable of recognizing, predicting, advising, optimizing,
  • bridging the gap between people and technology,
  • bridging the gap between people and people
• through the use neural networks,
  • leverged by many other AI / Stats tools
  • programmed in AI-developer-driven IDE such as RStudio
    • interfaced with Python, C++ (if needed)

Credits: NRC-CNRC, 2003

• http://collect-connect.cstmcweb.ca/2015/02/community-support-click-nose/
• https://www.nrc-cnrc.gc.ca/eng/achievements/highlights/2004/nouse_technology.html
• http://www.videorecognition.com/Nouse (http://www.Nouse.ca)

• http://www.videorecognition.com
• http://www.videorecognition.com/memory
• http://www.videorecognition.com/memory/pinn

… which we built in the past, and

… which we can built now and in future

URL: http://videorecognition.com/doc/presentations/2006-10-VRS-4VACE-briefed.pdf

Video Recognition Systems of NRC  

Project Leader:  Dr. Dmitry Gorodnichy

DNI IARPA VACE, Phase III Meeting 
Washington DC, 1 November 2006

• Dmitry O. Gorodnichy, Associative neural networks as means for low-resolution video-based recognition, International Joint Conference on Neural Networks (IJCNN'05), NRC 48217.

• "Best Presentation" award at IJCNN 1999

• D.O. Gorodnichy, A.M. Reznik, (1997) "Increasing Attraction of Pseudo-Inverse Autoassociative Networks" , Neural Processing Letters, volume 5, issue 2, pp. 123-127,

  • Compare to: Hochreiter, S.; Schmidhuber, J. (1997). "Long Short-Term Memory". Neural Computation. 9 (8): 1735–1780.
• Ph.D. Dissertation: "Investigation and Design of High Performance Fully-Connected Neural Networks", Glushkov Cybernetics Center, Ukraine, Kiev, 1997.
  • Ref: http://www.gorodnichy.ca/archives/phd/PINN

URL: http://videorecognition.com/doc/presentations/2005-AI-PavlovDogs.pdf, NRC 48209

Projection learning vs correlation learning: 
from Pavlov dogs to face recognition.  

Project Leader: D.O. Gorodnichy 

AI'05 Workshop on Correlation learning, 
Victoria, BC, Canada, 
May 8-12, 2005. 

Different terminology - same idea

• Neural networks, Neurocomputation, Connectionist models -> Deep Learning
• Image Processing / Computer Visions feature/edge detections –> Convolution networks
• Attractor-based Networks Neural –> Recurrent Neural Networks,
• Desaturation –> Regularization technique against over-fitting
• Flood-fill processing –> "Òne-hot" coding in Tensorflow
• All custom-made codes in MS C++ –> in R/RStudio with library(keras)

Now with much more efficient programming workflow and support !

• Revival of complex, fully-connected, reccurrent NN:
  • Propelled by Google, RStudio
  • Canada is doing well:
    • MILA (Bengio), UofT (Hinton), UofA (Sutton)
  • Many packages in R

• They are XXI century's "Scientific calculator" for everyone

• Trap: tendency to use them as "black box"
  • not enough preprocessing (thinking prior to running)
  • skipping understanding of the nature of the problem

• "Simply the Best!"
  • library(keras): it installs library(tensorflow)
    
• Leverages R/RStudio power and beauty:
  • pipeline %>% worksflow
    • most suitable for Tensor data flows,
  • RStudio IDE and web-servers, which are designed by AI scientists themselves
• Plenty of excellent free tutorials:
  • "Learn and apply Keras/Tensorflow in R/RStudio, Efficiently"

• MNIST digit recognition is used the "Hello world" example:

# Define model
model <- keras_model_sequential() %>%
  layer_conv_2d(filters = 32, kernel_size = c(3,3), activation = 'relu',
                input_shape = input_shape) %>%
  layer_conv_2d(filters = 64, kernel_size = c(3,3), activation = 'relu') %>%
  layer_max_pooling_2d(pool_size = c(2, 2)) %>% 
  layer_dropout(rate = 0.25) %>%
  layer_flatten() %>%
  layer_dense(units = 128, activation = 'relu') %>%
  layer_dropout(rate = 0.5) %>%
  layer_dense(units = 10, activation = 'softmax')
  
# Compile model  
model %>% compile(
  loss = loss_categorical_crossentropy,
  optimizer = optimizer_adadelta(),
  metrics = c('accuracy')
  
  
# Train model
model %>% fit(
  x_train, y_train, validation_split = 0.2,
  batch_size = batch_size, epochs = epochs 
)

• Extending to processing sequences of images:

...

Data from Open Government -> Historical Border Wait Times:

library(data.table);library(ggplot2);library(lubridate); 
library(magrittr); library(stringr);  
dt <- fread("http://cbsa-asfc.gc.ca/data/bwt-taf-2016-07-01--2016-09-30-en.csv")

Cleaning and selecting data:

dt <- dt[str_detect(Location, "BC")] # subset: BWT in British Columbia
dt[, Updated := as.POSIXct(Updated, format = "%Y-%m-%d %H:%M ")  ]
dt$BWT <- str_replace(dt$`Travellers Flow`, "No delay", "0") %>%  
  str_replace("Closed", "0") %>%   str_replace("Not applicable", "0") %>% 
  str_replace("Missed entry", "NA") %>% as.numeric()

dt <- dt[, .(Updated, BWT, Location)]

theme_set(theme_minimal())
ggplot(dt[Updated<dmy("01/09/2016")], aes(Updated,BWT,col=Location)) + 
  geom_step() + facet_grid(Location ~ .)

model2 <- keras_model_sequential() %>%
  layer_gru(units = 32, dropout = 0.2, recurrent_dropout = 0.2,
            input_shape = list(NULL, dim(dt)[[-1]])) %>%
  layer_dense(units = 1)

model2 %>% compile(
  optimizer = optimizer_rmsprop(),  loss = "mae"
)

model2 %>% fit_generator(
  train_gen,  steps_per_epoch = 500,  epochs = 40,
  validation_data = val_gen,  validation_steps = val_steps
)

Looking into the Future

What's common ?

This presentation available online: www.IVIM.ca/dg/ai

• Experimental beta versions:

  • https://bookdown.org/gorodnichy/ai1 (ioslides: more graphics, less text)
  • https://bookdown.org/gorodnichy/ai2 (Slidy: simpler layout, complete text)
  • https://bookdown.org/gorodnichy/ai3 (plain html)
  • https://bookdown.org/gorodnichy/ai4 (bookdown: includes epub and pdf)
  • https://bookdown.org/gorodnichy/ai-i (interactive with Shiny)

The RMarkdown/ R source of this document: https://raw.githubusercontent.com/gorodnichy/LA-Rmd-Shiny/master/D/pres_AI-NRC-2018.Rmd

Related:

• 'Learn and Apply interactive programming and presentations with RStudio Shiny'.
• 'Learn and Apply Text Analysis in R, Efficiently'

Contact: dg@ivim.ca (dmitry@gorodnichy.ca)