data analytics using deep learning
play

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY - PowerPoint PPT Presentation

Aug 16, 2023 •317 likes •1k views

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 1 : C O U R S E I N T R O D U C T I O N WELCOME TO 8803-DDL This is a cross-cutting course! Gain holistic understanding of three areas Data

  1. DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 1 : C O U R S E I N T R O D U C T I O N

  2. WELCOME TO 8803-DDL • This is a cross-cutting course! • Gain holistic understanding of three areas – Data Analytics – Machine Learning – Computer Vision • Bridge the gap between systems and machine learning GT 8803 // Fall 2019 2

  3. CREDITS • This course is derived from two courses • Convolutional Neural Networks for Visual Recognition – Fei Fei Li, Andrej Karpathy, and Justin Johnson – http://cs231n.stanford.edu/ • Advanced database systems – Andy Pavlo – https://15721.courses.cs.cmu.edu/ GT 8803 // Fall 2019 3

  4. TODAY’S AGENDA • Course Overview • Course Objectives • Course Logistics • History of Computer Vision • Visual Recognition Overview GT 8803 // Fall 2019 4

  5. COURSE OVERVIEW 5 GT 8803 // Fall 2018

  6. BIG DATA & DATA SCIENCE ERA • Visual data is the biggest Big Data out there Millions of images uploaded EVERY day Hours of videoS uploaded every minute GT 8803 // Fall 2019 6

  7. NEXT-GENERATION APPS • Apps will focus on visual data SELF-DRIVING CARS SPORTS ANALYTICS GT 8803 // Fall 2019 7

  8. CHALLENGES: TRADITIONAL DATABASE SYSTEMS • Traditional database systems only support structured data EMPLOYEE ID NAME AGE SALARY 101 PETER 25 100K 102 JOHN 20 80K 103 MARK 30 120K GT 8803 // Fall 2019 8

  9. WHY IS THIS IMPORTANT NOW? • Modern computer vision techniques have made great strides – Near human-levels of accuracy for several visual data analytics tasks GT 8803 // Fall 2019 9

  10. EXAMPLE: IMAGE CLASSIFICATION www.image-net.org 22 K categories and 15 M images Animals P lants • Structures • Person • Bird • Tree • Artifact • Scenes • Tools • Indoor • Fish • Flower • Mammal • Food • Appliances • GeologicalFormations • Invertebrate • Materials • Structures • SportActivities Deng, Dong, Socher, Li, Li, & Fei-Fei, 2009 GT 8803 // Fall 2019 10

  11. EXAMPLE: IMAGE CLASSIFICATION www.image-net.org OUTPUT: OUTPUT: Scale Scale T-shirt T-shirt Steel drum Giant Panda Drumstick Drumstick Mud turtle Mud turtle GT 8803 // Fall 2019 11

  12. EXAMPLE: IMAGE CLASSIFICATION www.image-net.org Russakovskyet al., IJCV2015 GT 8803 // Fall 2019 12

  13. CHALLENGES: DEEP LEARNING MODELS • Computational Efficiency • Usability GT 8803 // Fall 2019 13

  14. CHALLENGES: COMPUTER VISION PIPELINES • Computational Efficiency – These pipelines are computationally infeasible at scale – Example: State-of-the-art object detection models run at 3 frames per second (fps) (e.g., Mask R-CNN) – It will take 8 decades of GPU time to process 100 cameras over a month of video. GT 8803 // Fall 2019 14

  15. CHALLENGES: COMPUTER VISION PIPELINES • Usability – These techniques require complex, imperative programming across many low-level libraries (e.g., Pytorch and OpenCV) – This is an ad-hoc, tedious process that ignores opportunity for cross-operator optimization – Traditional database systems were successful due to their ease of use (i.e., SQL is declarative) GT 8803 // Fall 2019 15

  16. GOAL: DECLARATIVE VIDEO ANALYTICS SYSTEM GT 8803 // Fall 2019 16

  17. COURSE OBJECTIVES 17 GT 8803 // Fall 2018

  18. WHY SHOULD YOU TAKE THIS COURSE? • There are many challenging problems in database systems & machine learning • Systems + ML developers are in demand • If you are good enough to write code for a ML-driven data analytics system, then you can write code on almost anything else GT 8803 // Fall 2019 18

  19. COURSE OBJECTIVES • Learn about cutting-edge research topics in data analytics and deep learning • Learn about modern practices in systems programming and deep learning • We will cover state-of-the-art topics • This is not a course on classical database systems GT 8803 // Fall 2019 19

  20. PRE-REQUISITES • Proficiency in Python and some high-level familiarity with C++ – All assignments will be in Python; but some of the deep learning libraries we may look at later in the class will be written in C++ – A Python tutorial is available on course website • Calculus, Linear Algebra • Basic Probability and Statistics GT 8803 // Fall 2019 20

  21. PRE-REQUISITES • Fundamentals of Machine Learning – We will be formulating cost functions, taking derivatives and performing optimization with gradient descent • I am happy to have people from different backgrounds – But talk to me if you’re not sure GT 8803 // Fall 2019 21

  22. COURSE LOGISTICS 22 GT 8803 // Fall 2018

  23. COURSE LOGISTICS • Office: Klaus 3324 • On-line Discussion through Piazza: – https://piazza.com/gatech/fall2019/cs8803ddl/home • For all technical questions, please use Piazza – Don’t email me directly – All non-technical questions should be sent to me GT 8803 // Fall 2019 23

  24. COURSE LOGISTICS • Course Schedule – https://www.cc.gatech.edu/~jarulraj/courses/8803 -f19/pages/schedule.html – We will post lecture slides and course materials on this page • Course Policies – Students are expected to abide by the Georgia Tech Honor Code – If you are not sure, ask me GT 8803 // Fall 2019 24

  25. COURSE LOGISTICS • Grading Tool: Gradescope – https://www.gradescope.com/courses/54455 – You will get immediate feedback on your programming assignments – You can iteratively improve your score over time GT 8803 // Fall 2019 25

  26. GRADE BREAKDOWN • The final grade for the course will be tentatively based on the following weights: – 30% Assignments – 30% Midterm Exam – 40% Group Project • Emphasis on learning rather than testing you – If your project is truly amazing, you get an automatic A! GT 8803 // Fall 2019 26

  27. TEACHING ASSISTANTS • TA #1: Jaeho Bang – Ph.D. student in Computer Science – B.S. from Carnegie Mellon • TA #2: TBD GT 8803 // Fall 2019 27

  28. OFFICE HOURS • Immediately before class – Me: Mon/Wed 3:30 – 4:30 PM – Jaeho: Tue/Thu 3:30 – 4:30 PM – Near my office (Klaus 3324) • Things we can talk about – Questions related to lectures and assignments – Project ideas – Can’t give relationship advice GT 8803 // Fall 2019 28

  29. HISTORY OF COMPUTER VISION 29 GT 8803 // Fall 2018

  30. EVOLUTION’s BIG BANG • ~543 million years – What was life like back then? – Onset of vision triggered evolution’s Big Bang – Now biggest sensory system in most animals GT 8803 // Fall 2019 30

  31. CAMERA OBSCURA GEMMA FRISIUS (1545) DA VINCI (~1500) ENCYCLOPEDIE (~1800) 31 GT 8803 // Fall 2019

  32. ELECTROPHYSIOLOGY (1959) • Visual processing mechanism in mammals Electricalsignal Simple cells : frombrain Response to light orientation Complexcells : Response to light orientation and movement Hypercomplex cells : Responseto movement with an endpoint Stimulus Stimulus Response Noresponse Response (end point) GT 8803 // Fall 2019

  33. BLOCK WORLD (1961) • Visual world simplified into geometric shapes (c) Feature points selected (b) Differentiated picture (a) Original picture GT 8803 // Fall 2019 33

  34. PROJECT MAC (1966) GT 8803 // Fall 2019 34

  35. STAGES OF VISUAL REPRESENTATION (1970 s ) INPUT IMAGE EDGE IMAGE 2.5-D MODEL 3-D MODEL Local surface 3-D models Zero crossings, hierarchically orientation and Perceived edges, bars, organized in discontinuities in Intensities ends, virtual depth and terms of surface lines, groups, and volumetric surface curves orientation primitives boundaries GT 8803 // Fall 2019 35

  36. BETTER REPRESENTATIONS (1970 s ) GENERALIZED PICTORIAL CYLINDER STRUCTURE (1979) (1973) GT 8803 // Fall 2019 36

  37. OBJECT RECOGNITION (1987) GT 8803 // Fall 2019 37

  38. IMAGE SEGMENTATION (1987) GT 8803 // Fall 2019 38

  39. FACE DETECTION (2001) GT 8803 // Fall 2019 39

  40. FEATURE-BASED OBJECT RECOGNITION (1999) • Certain features are invariant to perspective GT 8803 // Fall 2019 40

  41. FEATURE MATCHING (2006) SPATIAL PYRAMID LEVEL 1 LEVEL 0 GT 8803 // Fall 2019 41

  42. HUMAN POSE DETECTION (2005) frequency GT 8803 // Fall 2019 42

  43. PASCAL VISUAL OBJECT CHALLENGE (2006~12) Train Person Airplane 20 OBJECT CATEGORIES GT 8803 // Fall 2019 43

  44. IMAGENET CHALLENGE (2009~17) www.image-net.org 22K categories and 15M images Animals P lants • Structures • Person • Bird • Tree • Artifact • Scenes • Tools • Indoor • Fish • Flower • Mammal • Food • Appliances • GeologicalFormations • Invertebrate • Materials • Structures • SportActivities Deng, Dong, Socher, Li, Li, & Fei-Fei, 2009 GT 8803 // Fall 2019 44

  45. IMAGENET CHALLENGE (2009~17) www.image-net.org OUTPUT: OUTPUT: Scale Scale T-shirt T-shirt Steel drum Giant Panda Drumstick Drumstick Mud turtle Mud turtle GT 8803 // Fall 2019 45

  46. IMAGENET CHALLENGE (2009~17) www.image-net.org Russakovskyet al., IJCV2015 GT 8803 // Fall 2019 46

  47. VISUAL RECOGNITION OVERVIEW 47 GT 8803 // Fall 2018

  48. IMAGE CLASSIFICATION • This course will focus on one of the most fundamental problems of visual recognition – Image classification • This technique can be applied in many ways GT 8803 // Fall 2019 48

Recommend

DATA ANALYTICS USING DEEP LEARNING GT 8803 FALL 2018 POOJA BHANDARY T O W A R D S A U N I F

DATA ANALYTICS USING DEEP LEARNING GT 8803 FALL 2018 POOJA BHANDARY T O W A R D S A U N I F

DATA ANALYTICS USING DEEP LEARNING GT 8803 FALL 2018 POOJA BHANDARY T O W A R D S A U N I F I E D A R C H I T E C T U R E F O R I N - R D B M S A N A L Y T I C S TODAYs PAPER SIGMOD 2012 In-RDBMS Analytics Hazy project at

317 views • 31 slides
DATA ANALYTICS USING DEEP LEARNING A U T O M A T I C D A T A B A S E M A N A G E M E N T S Y S

DATA ANALYTICS USING DEEP LEARNING A U T O M A T I C D A T A B A S E M A N A G E M E N T S Y S

DATA ANALYTICS USING DEEP LEARNING A U T O M A T I C D A T A B A S E M A N A G E M E N T S Y S T E M T U N I N G T H R O U G H L A R G E - S C A L E M A C H I N E L E A R N I N G S I D D H A R T H B I S W A L TODAYs PAPER

496 views • 32 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 8 : Q

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 8 : Q

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 8 : Q U E R Y E X E C U T I O N administrivia Reminders Sign up for discussion slots on Thursday Proposal presentations on next Wednesday

1.32k views • 112 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 9 : Q

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 9 : Q

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 9 : Q U E R Y O P T I M I Z A T I O N administrivia Reminders Assignment 1: postponed to next Monday Sign up for discussion slots on Thursday

1.68k views • 150 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 6 : D

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 6 : D

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 6 : D I S K - C E N T R I C A N D I N - M E M O R Y D A T A B A S E S Y S T E M S administrivia Project ideas List shared on Piazza Start

1.22k views • 80 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 1 2 : T

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 1 2 : T

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 1 2 : T R A I N I N G N E U R A L N E T W O R K S ( P T 1 ) administrivia Reminders Integration with Eva Code reviews Each team must send

934 views • 70 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 7 : S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 7 : S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 0 7 : S T O R A G E M O D E L S & C O M P R E S S I O N administrivia Reminder Assignment 1 due on next Wednesday Sign up for discussion

1.09k views • 63 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // SIDDHARTH BISWAL L E C T U R E # 0 3 : B L A Z E

DATA ANALYTICS USING DEEP LEARNING GT 8803 // SIDDHARTH BISWAL L E C T U R E # 0 3 : B L A Z E

DATA ANALYTICS USING DEEP LEARNING GT 8803 // SIDDHARTH BISWAL L E C T U R E # 0 3 : B L A Z E I T : F A S T E X P L O R A T O R Y V I D E O Q U E R I E S U S I N G N E U R A L N E T W O R K S TODAYs PAPER BlazeIt: Fast

280 views • 23 slides
Effectiveness of Deep Learning Vs. Machine Learning in a Health Care Use Case RxToDx A Data

Effectiveness of Deep Learning Vs. Machine Learning in a Health Care Use Case RxToDx A Data

Effectiveness of Deep Learning Vs. Machine Learning in a Health Care Use Case RxToDx A Data Science Machine Learning/Deep Learning Show Case Dima Rekesh/Julie Zhu/Ravi Rajagopalan Optum Technology November, 2017 Analytics Machine

483 views • 32 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ S P E A K I N G T I P S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ S P E A K I N G T I P S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ S P E A K I N G T I P S CREDITS Based on a talk given by: Margaret Martonosi (Princeton) Computer architect GT 8803 // Fall 2019 2 MOTIVATION

871 views • 75 slides
UNIFIED MEMORY FOR DATA ANALYTICS AND DEEP LEARNING Nikolay Sakharnykh, Chirayu Garg, and Dmitri

UNIFIED MEMORY FOR DATA ANALYTICS AND DEEP LEARNING Nikolay Sakharnykh, Chirayu Garg, and Dmitri

UNIFIED MEMORY FOR DATA ANALYTICS AND DEEP LEARNING Nikolay Sakharnykh, Chirayu Garg, and Dmitri Vainbrand, Thu Mar 19, 3:00 PM RAPIDS CUDA-accelerated Data Science Libraries PYTHON DL RAPIDS FRAMEWORKS DASK / SPARK cuGraph cuDF cuML

720 views • 46 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JACOB LOGAS L E C T U R E # 1 0 : L

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JACOB LOGAS L E C T U R E # 1 0 : L

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JACOB LOGAS L E C T U R E # 1 0 : L O C A L I T Y - S E N S I T I V E H A S H I N G F O R E A R T H Q U A K E D E T E C T I O N TODAYS PAPER Locality-Sensitive Hashing for

458 views • 41 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 2 0 : A

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 2 0 : A

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 2 0 : A D V E R S A R I A L T R A I N I N G administrivia Reminders Best project prize Quiz cancelled Guest lecture GT 8803 // Fall 2019

1.1k views • 47 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 3 : F

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 3 : F

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 3 : F O C U S : Q U E R Y I N G L A R G E V I D E O D A T A S E T S W I T H L O W L A T E N C Y A N D L O W C O S T TODAYS PAPER Focus:

1.06k views • 56 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 4 : L I

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 4 : L I

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // JENNIFER MA L E C T U R E # 1 4 : L I V E V I D E O A N A L Y T I C S A T S C A L E W I T H A P P R O X I M A T I O N A N D D E L A Y - T O L E R A N C E TODAYS PAPER Live

709 views • 56 slides
AI and Predictive Analytics in Data-Center Environments Neural Networks and Deep Learning Josep

AI and Predictive Analytics in Data-Center Environments Neural Networks and Deep Learning Josep

AI and Predictive Analytics in Data-Center Environments Neural Networks and Deep Learning Josep Ll. Berral @BSC Intel Academic Education Mindshare Initiative for AI Introduction Neural networks attempt to imitate the brain neuron

339 views • 21 slides
DSC 102 Systems for Scalable Analytics Arun Kumar Topic 6: Deep Learning Systems 1 Outline

DSC 102 Systems for Scalable Analytics Arun Kumar Topic 6: Deep Learning Systems 1 Outline

DSC 102 Systems for Scalable Analytics Arun Kumar Topic 6: Deep Learning Systems 1 Outline Rise of Deep Learning Methods Deep Learning Systems: Specification Deep Learning Systems: Execution Future of Deep Learning Systems

561 views • 25 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // CHRISTINE HERLIHY L E C T U R E # 0

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // CHRISTINE HERLIHY L E C T U R E # 0

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // CHRISTINE HERLIHY L E C T U R E # 0 8 : T E N S O R F L O W : A S Y S T E M F O R L A R G E - S C A L E M A C H I N E L E A R N I N G TODAYS PAPER TensorFlow: A system

451 views • 27 slides
Big Data Analytics 1. Rapid development: HEP at new level of Big Data Analysis 2. Education:

Big Data Analytics 1. Rapid development: HEP at new level of Big Data Analysis 2. Education:

deep learning in particle & astroparticle physics Big Data Analytics 1. Rapid development: HEP at new level of Big Data Analysis 2. Education: Anchoring machine learning in curriculum 3. Funding: Competence networks of various disciplines

542 views • 24 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ W R I T I N G T I P S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ W R I T I N G T I P S

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ W R I T I N G T I P S ANALYSIS 2 GT 8803 // Fall 2018 ANALYSIS Problem Description Significance Novelty Relevance Validity Contribution GT 8803

663 views • 33 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // VENKATA KISHORE PATCHA Lecture#16 :

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // VENKATA KISHORE PATCHA Lecture#16 :

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // VENKATA KISHORE PATCHA Lecture#16 : In-RDBMS Hardware Acceleration of Advanced Analytics TODAYS PAPER In- RDBMS Hardware Acceleration of Advanced Analytics Authors:

558 views • 40 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // NIDHI MENON LECTURE #15: THE DATA

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // NIDHI MENON LECTURE #15: THE DATA

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2018 // NIDHI MENON LECTURE #15: THE DATA CALCULATOR: DATA STRUCTURE DESIGN AND COST SYNTHESIS FROM FIRST PRINCIPLES TODAYs PAPER The Data Calculator: Data Structure Design and Cost

522 views • 39 slides
Deep learning for retail analytics and reference data management Alessandro Zolla Robert Bogucki

Deep learning for retail analytics and reference data management Alessandro Zolla Robert Bogucki

Deep learning for retail analytics and reference data management Alessandro Zolla Robert Bogucki Nielsen Scope Nielsen measures what people... WATCH BUY TV Ratings Brick & Mortar Advertising exposure eCommerce TV

531 views • 22 slides
DATA ANALYTICS USING DEEP LEARNING GT 8803 // VENKATA KISHORE PATCHA L E C T U R E # 0 6 : S M

DATA ANALYTICS USING DEEP LEARNING GT 8803 // VENKATA KISHORE PATCHA L E C T U R E # 0 6 : S M

DATA ANALYTICS USING DEEP LEARNING GT 8803 // VENKATA KISHORE PATCHA L E C T U R E # 0 6 : S M E L L Y R E L A T I O N S : M E A S U R I N G A N D U N D E R S T A N D I N G D A T A B A S E S C H E M A Q U A L I T Y TODAYS PAPER

394 views • 24 slides

More recommend