deep learning

https://machinelearningmastery.com/configure-encoder-decoder-model-neural-machine-translation/Intel

http://pytorch.org/

Intel AI Academy : https://software.intel.com/en-us/ai-academy

Access Instructional Material by applying on the link below :

Intel AI Academy for Professors - https://software.intel.com/en-us/ai-academy/professors

Intel Student Ambassador Program - https://software.intel.com/en-us/ai-academy/ambassadors

Intel Student Kits - https://software.intel.com/en-us/ai-academy/students/kits

Intel Software Tools and SDKs - https://software.intel.com/en-us/ai-academy/tools

Intel Dev Cloud Access Request - https://software.intel.com/en-us/ai-academy/tools/devcloud

GAN

Conditional Generative Adversarial Nets by Mehdi Mirza https://arxiv.org/pdf/1411.1784.pdf

Image-to-Image Translation with Conditional Adversarial Networks Phillip Isola Jun-Yan Zhu Tinghui Zhou Alexei A. Efros https://arxiv.org/pdf/1611.07004.pdf

Generative Adversarial Nets Ian J. Goodfellow, Jean Pouget-Abadie∗ , Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair† , Aaron Courville, Yoshua Bengio‡ https://arxiv.org/pdf/1406.2661.pdf

machine translation

• • https://machinelearningmastery.com/configure-encoder-decoder-model-neural-machine-translation/

  • Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation https://arxiv.org/abs/1609.08144

  • Improved Neural Machine Translation with SMT Features. (English to Chinese paper handle unk, out of vocabulary problem oov)

  • https://chunml.github.io/ChunML.github.io/project/Sequence-To-Sequence/

  • https://pdfs.semanticscholar.org/2701/f0782782ce409473a26432b8cef7e4b224d0.pdf (MSC thesis: Towards efficient Neural Machine Translation for Indian Languages)

  • word embedding https://en.wikipedia.org/wiki/Word2vec

    • • Linguistic Regularities in Continuous Space Word Representations : http://www.aclweb.org/anthology/N13-1090, http://www.isca-speech.org/archive/archive_papers/interspeech_2010/i10_1045.pdf

      • tutorial on word embedding : http://mccormickml.com/2016/04/19/word2vec-tutorial-the-skip-gram-model/

  • https://towardsdatascience.com/word2vec-skip-gram-model-part-1-intuition-78614e4d6e0b

Ne

• https://www.quantamagazine.org/new-theory-cracks-open-the-black-box-of-deep-learning-20170921/

capsule network

• Hinton

  • (MATRIX CAPSULES WITH EM ROUTING ) https://openreview.net/pdf?id=HJWLfGWRb

    • (Dynamic Routing Between Capsules) https://arxiv.org/pdf/1710.09829.pdf

• https://hackernoon.com/what-is-a-capsnet-or-capsule-network-2bfbe48769cc

• https://hackernoon.com/capsule-networks-are-shaking-up-ai-heres-how-to-use-them-c233a0971952

SFM using Neural net

3D-R2N2: Single or Multiple View 3D Object reconstruction https://arxiv.org/pdf/1604.00449.pdf

SFM net: learning structure and motion from video https://arxiv.org/pdf/1704.07804.pdf

SE3-net: Learning Rigid Body Motion using Deep Neural Networks https://arxiv.org/pdf/1606.02378.pdf

IMU and Kalman

1. https://www.kudan.eu/kudan-news/an-introduction-to-slam/ 2016 with videos, a good introduction to slam

2. http://www.ece.ust.hk/~eeshaojie/ismar2017peiliang.pdf (Monocular Visual-Inertial State Estimation for Mobile Augmented Reality) 2017, HKUST, slam sfm+ Karlamn

3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3649364/ (Extended Kalman Filter-Based Methods for Pose Estimation Using Visual, Inertial and Magnetic Sensors 2013, pose using checker board)

4. http://appliedmaths.sun.ac.za/~wbrink/students/LHughes2014.pdf (Enhancing Mobile Camera Pose Estimation Through the Inclusion of Sensors, msc thesis, pose using checker board, p79)

5. http://openaccess.thecvf.com/content_cvpr_workshops_2015/W12/papers/Tiefenbacher_Off-the-Shelf_Sensor_Integration_2015_CVPR_paper.pdf 9using passive markers for pose estimation, UKF)

   1. http://www.mdpi.com/1424-8220/17/10/2164/htm (Pose Estimation of a Mobile Robot Based on Fusion of IMU Data and Vision Data Using an Extended Kalman Filter, sensor 2017, cityu HK, object recognition known model (section 3.2)., EKF, ransac)object recognition, known model,EKF

6. https://arxiv.org/ftp/arxiv/papers/1411/1411.2335.pdf (AN IMPROVED TRACKING USING IMU AND VISION FUSION FOR MOBILE AUGMENTED REALITY APPLICATIONS 2014) EKF, CAD known model to get pose , mixed with IMU complimentary filter

7. https://link.springer.com/content/pdf/10.1007%2Fs10846-010-9490-z.pdf (Fusion of IMU and Vision for Absolute Scale Estimation in Monocular SLAM 2011) Klein and Murray’s SLAM algorithm- bundle adjustment, plus IMU (different sampling scheme)

8. http://www.control.isy.liu.se/research/reports/LicentiateThesis/Lic1370.pdf (Pose Estimation and Calibration Algorithms for Vision and Inertial Sensors 2008, phd thesis) p12, may be known model

9. http://www.mit.edu/~shayegan/files/vision_based_pose_estimation_of_quads.pdf (Vision-Based Pose Estimation of Quadcopters using UKFs2014) posit, knwon model, ukf for Quadcopters

10. https://venturi.fbk.eu/wp-content/uploads/2011/10/PorRicCia_EESMS_2012.pdf (Visual-inertial Tracking on Android for Augmented Reality Applications 2012 )vision using G. Klein and D. Murray, “Parallel tracking and mapping on a cameraphone,” in Proc. Eigth IEEE and ACM International Symposium on Mixed and Augmented Reality (ISMAR’09), 2009

11. http://jultika.oulu.fi/files/nbnfioulu-201609142782.pdf (POSE ESTIMATION USING TWO LINE CORRESPONDENCES AND GRAVITY VECTOR FOR IMAGE RECTIFICATION msc thesis 2016) line correspondences

• 1. khwong

http://www.cse.cuhk.edu.hk/%7Ekhwong/j2004_IEEE_chang_MM_xlowe_draft.pdf

http://www.cse.cuhk.edu.hk/%7Ekhwong/j2004_IEEE_yu_SMC_B_kalman_draft.pdf

http://www.cse.cuhk.edu.hk/~khwong/c174_pt_line_trifocal_kalman.pdf

http://www.cse.cuhk.edu.hk/~khwong/c174_pt_line_trifocal_kalman.pdfPose estaimtion

POSE estimation

http://home.in.tum.de/~grembowi/ar2004_05/3dPoseEstimation_presentation.pdf

POSIT

• • https://pdfs.semanticscholar.org/8dc2/67b5e0e83aa293f1595482ff52547db808f3.pdf the original paper ModelBased Object Pose in 25 Lines of Code

  • A Comparison of Iterative 2D-3D Pose Estimation Methods for Real-Time Applications http://projekter.aau.dk/projekter/files/14427578/A_Comparison_of_2D-3D_Pose_Estimation_Methods.pdf

  • https://link.springer.com/content/pdf/10.1007%2F978-3-642-02230-2.pdf

  • http://home.in.tum.de/~grembowi/ar2004_05/3dPoseEstimation_elaboration.pdf a tutorial on pose estimation method including posit

  • http://www.haowuhw.com/pose/report.pdf a report

  • code

    • http://people.rennes.inria.fr/Eric.Marchand/pose-estimation/tutorial-pose-dementhon-opencv.html

    • http://www.cfar.umd.edu/~daniel/Site_2/Code.html (poseit matlab , c++ etc)

dataset

http://www.lrec-conf.org/proceedings/lrec2014/pdf/774_Paper.pdf

data science

big-data-35-brilliant-and-free-data-sources

https://www.analyticsvidhya.com/blog/2016/10/17-ultimate-data-science-projects-to-boost-your-knowledge-and-skills/

text mining

http://tidytextmining.com/sentiment.html

Demo

face track by C. kam

Current work

• http://opus.lingfil.uu.se/TED2013.php

  • Vuforia VR/AR : https://developer.vuforia.com/

  • aruco markers: https://www.youtube.com/watch?v=dNKHZW1qLCs

Deep Learning and Kalman filtering

Deep Learning Book. http://www.deeplearningbook.org/

Kalman matlab toolbox Aalto University:EKF/UKF Toolbox for Matlab V1.3, documentation.pdf, http://becs.aalto.fi/en/research/bayes/ekfukf/

>download and run \ekfukf-master\demos\bot_demo, comment , set silent =1 in bot_demo_all.m or uks_bot_dem.m. remove pause in bot_h.m

1. 3-D and RNN http://3d-r2n2.stanford.edu/ 2. SLAM and RNN https://arxiv.org/pdf/1701.08376.pdf 3. Bayesian filter and RNN http://research.nvidia.com/sites/default/files/pubs/2017-07_Dynamic-Facial-Analysis/rnnface.pdf

Windows: How to use tensor flow in PC-windows, Install Mobaxterm, so your pc has xterm to use the GPU machine xxxyy

$CUDA_VISIBLE_DEVCES=5;6;7 tensorflow minst_cnn.py

Editing $subl file.py %sublime for xterm editing Or $vim file.py # text mode only

Linux:$ssh –XC xxxxyy.cse.cuhk.edu.hk, To use tensorflow as usual

• • paper studying

    • 1st week from 7 Aug 2017

      • http://www.deeplearningbook.org/ (ch.4)

    • 2nd week from 14 July 2017

      • http://www.aclweb.org/anthology/N16-1030 (LSTM: neural architecture for named entity regression)

      • http://yann.lecun.com/exdb/publis/pdf/farabet-pami-13.pdf (learning hierarchical features for scene labeling)

      • https://people.cs.umass.edu/~mccallum/papers/crf-tutorial.pdf Conditional random field

  • 3rd week from 21 Aug 2017

    • Ch5 of http://www.deeplearningbook.org/

Papers:

Fully convolutional networks for semantic segmentation by J Long, E Shelhamer, T Darrell

Sequence to sequence learning with neural networks by I Sutskever, O Vinyals, QV Le -

tutorials

http://colah.github.io/posts/2015-08-Understanding-LSTMs/

https://github.com/terryum/awesome-deep-learning-papers

Data

• • http://archive.ics.uci.edu/ml/index.php (UC Irvine Machine Learning Repository)

tutorial:

https://theneuralperspective.com/tag/tutorials/

• rnn-encoder-decoder

  • • http://anie.me/rnn-encoder-decoder/ a good guide

• • text generation using lstm

    • https://chunml.github.io/ChunML.github.io/project/Creating-Text-Generator-Using-Recurrent-Neural-Network/ (by Trung Tran)

  • Backpropagagtion

    • https://mattmazur.com/2015/03/17/a-step-by-step-backpropagation-example/ (good numerical examples)

    • https://medium.com/@erikhallstrm/backpropagation-from-the-beginning-77356edf427d good backpro tutorial ,

  • RNN encoder-decoder

• https://theneuralperspective.com/2016/11/20/recurrent-neural-networks-rnn-part-3-encoder-decoder/

• • LSTM tutorial ppt

    • http://www.iro.umontreal.ca/~pift6266/A07/documents/lstm.pdf (LSTM for speech and music ,with good diagrams describing the functions of gates)

  • LSTM papers

    • http://www.bioinf.jku.at/publications/older/2604.pdf (the original 1997 paper)

    • http://www.felixgers.de/papers/phd.pdf (the forget gate thesis, 2001)

    • https://arxiv.org/pdf/1503.04069.pdf (search space odyssey)

    • http://karpathy.github.io/2015/05/21/rnn-effectiveness/ (good) The Unreasonable Effectiveness of Recurrent Neural Networks

    • http://blog.echen.me/2017/05/30/exploring-lstms/ by Edwin Chen

  • softmax (cross-entropy loss)

    • https://www.quora.com/Is-the-softmax-loss-the-same-as-the-cross-entropy-loss (difference between softmax loss and cross-entropy loss)

    • "However, people use the term "softmax loss" when referring to "cross-entropy loss" and because you know what they mean, there's no reason to annoyingly correct them."

    • https://stats.stackexchange.com/questions/79454/softmax-layer-in-a-neural-network stated clearly the fomulars for backprog

    • https://www.ics.uci.edu/~pjsadows/notes.pdf (full derivation and include back propagation formulas)

  • http://peterroelants.github.io/posts/neural_network_implementation_intermezzo02/

  • https://datascience.stackexchange.com/questions/9302/the-cross-entropy-error-function-in-neural-networks

    • https://stackoverflow.com/questions/24351206/backpropagation-for-rectified-linear-unit-activation-with-cross-entropy-error (matlab)

  • vanishing gradient proboem

  • https://ayearofai.com/rohan-4-the-vanishing-gradient-problem-ec68f76ffb9b

  • sequence to sequence model

  • https://arxiv.org/pdf/1703.01619.pdf

  • https://indico.io/blog/sequence-modeling-neuralnets-part1/

  • https://medium.com/towards-data-science/lstm-by-example-using-tensorflow-feb0c1968537

  • https://google.github.io/seq2seq/nmt/

    • https://chunml.github.io/ChunML.github.io/project/Sequence-To-Sequence/ (by Trung Tran)

    • https://github.com/farizrahman4u/seq2seq code-theory iKeras

    • https://medium.com/@ageitgey/machine-learning-is-fun-part-5-language-translation-with-deep-learning-and-the-magic-of-sequences-2ace0acca0aa (good tutorial and numerical example )

    • attention

      • https://arxiv.org/pdf/1409.0473.pdf

    • parameters of lstm

    • https://stackoverflow.com/questions/38080035/how-to-calculate-the-number-of-parameters-of-an-lstm-network

    • https://datascience.stackexchange.com/questions/10615/number-of-parameters-in-an-lstm-model

    • https://stackoverflow.com/questions/38080035/how-to-calculate-the-number-of-parameters-of-an-lstm-network

    • https://www.quora.com/What-is-the-meaning-of-%E2%80%9CThe-number-of-units-in-the-LSTM-cell

    • https://www.quora.com/In-LSTM-how-do-you-figure-out-what-size-the-weights-are-supposed-to-be

      • http://kbullaughey.github.io/lstm-play/lstm/ (batch size example)

    • feedback (back prop)

    • https://medium.com/@aidangomez/let-s-do-this-f9b699de31d9

      • http://neuralnetworksanddeeplearning.com/chap2.html (How the backpropagation algorithm works)

    • Numerical examples

    • https://blog.aidangomez.ca/2016/04/17/Backpropogating-an-LSTM-A-Numerical-Example/

    • https://karanalytics.wordpress.com/2017/06/06/sequence-modelling-using-deep-learning/

    • http://monik.in/a-noobs-guide-to-implementing-rnn-lstm-using-tensorflow/

    • visualize LSTM

    • http://kvitajakub.github.io/2016/04/14/rnn-diagrams/

    • LSTM code

      • http://nicodjimenez.github.io/2014/08/08/lstm.html S9imple LSTM python-code , formulas and theory)

      • http://blog.csdn.net/u010866505/article/details/74910525 (MATLAB)

      • https://iamtrask.github.io/2015/11/15/anyone-can-code-lstm/ (python)

      • http://machinelearningmastery.com/how-to-use-an-encoder-decoder-lstm-to-echo-sequences-of-random-integers/ Keras: How to use an Encoder-Decoder LSTM to Echo Sequences of Random Integers

    • LSTM RNN tutorial

      • https://ayearofai.com/rohan-lenny-3-recurrent-neural-networks-10300100899b very detail

    • tutorial

    • http://people.idsia.ch/~juergen/lstm2003tutorial.pdf

      • https://zybuluo.com/hanbingtao/note/581764 (in Chinese)

    • Thesis

    • http://www.felixgers.de/papers/phd.pdf

    • The vanishing gradient problem

    • http://neuralnetworksanddeeplearning.com/chap5.html

    • https://www.quora.com/How-does-LSTM-help-prevent-the-vanishing-and-exploding-gradient-problem-in-a-recurrent-neural-network

    • GPU

      • Some information about the external GPU, please check the following link.

      • https://egpu.io/setup-guide-external-graphics-card-mac/#tb3-enclosures

      • https://www.hksilicon.com/articles/1028524