mle part 2
play

MLE part 2 18 / 70 Gaussian Mixture Model Suppose data is drawn - PowerPoint PPT Presentation

Mar 02, 2023 •266 likes •1.67k views

MLE part 2 18 / 70 Gaussian Mixture Model Suppose data is drawn from k Gaussians, meaning Y = j Discrete ( 1 , . . . , k ) , X = x | Y = j N ( j , j ) , and the parameters are = (( 1 , 1 , 1 ) , . . . , ( k ,

  1. Latent variable models. k p θ ( y i = j | x i ) and p θ ( y i = j | x i ) = p θ ( y i = j, x i ) � Since 1 = , then p θ ( x i ) j =1 n n n k � � � � L ( θ ) = ln p θ ( x i ) = 1 · ln p θ ( x i ) = p θ ( y i = j | x i ) ln p θ ( x i ) i =1 i =1 i =1 j =1 n k p θ ( y i = j | x i ) ln p θ ( x i , y i = j ) � � = p θ ( y i = j | x i ) . i =1 j =1 Therefore: define augmented likelihood n k R ij ln p θ ( x i , y i = j ) � � L ( θ ; R ) := ; R ij i =1 j =1 note that R ij := p θ ( y i = j | x i ) implies L ( θ ; R ) = L ( θ ) . 29 / 70

  2. E-M method for latent variable models Define augmented likelihood n k R ij ln p θ ( x i , y i = j ) � � L ( θ ; R ) := , R ij i =1 j =1 with responsibility matrix R ∈ R n,k := { R ∈ [0 , 1] n × k : R 1 k = 1 n } . Alternate two steps: ◮ E-step: set ( R t ) ij := p θ t − 1 ( y i = j | x i ) . ◮ M-step: set θ t = arg max θ ∈ Θ L ( θ ; R t ) . 30 / 70

  3. E-M method for latent variable models Define augmented likelihood n k R ij ln p θ ( x i , y i = j ) � � L ( θ ; R ) := , R ij i =1 j =1 with responsibility matrix R ∈ R n,k := { R ∈ [0 , 1] n × k : R 1 k = 1 n } . Alternate two steps: ◮ E-step: set ( R t ) ij := p θ t − 1 ( y i = j | x i ) . ◮ M-step: set θ t = arg max θ ∈ Θ L ( θ ; R t ) . Soon: we’ll see this gives nondecreasing likelihood! 30 / 70

  4. E-M for Gaussian mixtures Initialization: a standard choice is π j = 1 / k , Σ j = I , and ( µ j ) k j =1 given by k -means. ◮ E-step: Set R ij = p θ ( y i = j | x i ) , meaning π j p µ j , Σ j ( x i ) R ij = p θ ( y i = j | x i ) = p θ ( y i = j, x i ) = . � k p θ ( x i ) l =1 π l p µ l , Σ l ( x i ) ◮ M-step: solve arg max θ ∈ Θ L ( θ ; R ) , meaning � n � n i =1 R ij i =1 R ij π j := = , � n � k n l =1 R il i =1 � n � n i =1 R ij x i i =1 R ij x i µ j := = , � n nπ j i =1 R ij � n i =1 R ij ( x i − µ j )( x i − µ j ) T Σ j := . nπ j (These are as before.) 31 / 70

  5. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  6. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  7. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  8. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  9. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  10. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  11. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  12. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  13. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  14. Demo: spherical clusters 6 4 2 0 2 2 0 2 4 6 8 10 (Initialized with k -means, thus not so dramatic.) 32 / 70

  15. Demo: elliptical clusters E. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  16. Demo: elliptical clusters E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  17. Demo: elliptical clusters E. . . M. . . E. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  18. Demo: elliptical clusters E. . . M. . . E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  19. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  20. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  21. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  22. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  23. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  24. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 33 / 70

  25. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  26. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  27. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  28. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  29. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  30. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  31. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  32. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  33. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  34. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  35. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  36. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  37. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  38. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  39. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  40. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  41. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  42. Demo: elliptical clusters E. . . M. . . E. . . M. . . E. . . M. . . E. . . M. . . 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 33 / 70

  43. Theorem. Suppose ( R 0 , θ 0 ) ∈ R n,k × Θ arbitrary, thereafter ( R t , θ t ) given by E-M: ( R t ) ij := p θ t − 1 ( y = j | x i ) . and θ t := arg max L ( θ ; R t ) θ ∈ Θ Then L ( θ t ; R t ) ≤ R ∈R n × k L ( θ t ; R ) = L ( θ t ; R t +1 ) = L ( θ t ) max ≤ L ( θ t +1 ; R t +1 ) . In particular, L ( θ t ) ≤ L ( θ t +1 ) . 34 / 70

  44. Theorem. Suppose ( R 0 , θ 0 ) ∈ R n,k × Θ arbitrary, thereafter ( R t , θ t ) given by E-M: ( R t ) ij := p θ t − 1 ( y = j | x i ) . and θ t := arg max L ( θ ; R t ) θ ∈ Θ Then L ( θ t ; R t ) ≤ R ∈R n × k L ( θ t ; R ) = L ( θ t ; R t +1 ) = L ( θ t ) max ≤ L ( θ t +1 ; R t +1 ) . In particular, L ( θ t ) ≤ L ( θ t +1 ) . Remarks. ◮ We proved a similar guarantee for k -means, which is also an alternating minimization scheme. ◮ Similarly, MLE for Gaussian mixtures is NP-hard; it is also known to need exponentially many samples in k to information-theoretically recover the parameters. 34 / 70

  45. Proof. We’ve already shown: ◮ L ( θ t ; R t +1 ) = L ( θ t ) ; ◮ L ( θ t ; R t +1 ) = max θ ∈ Θ L ( θ ; R t +1 ) ≤ L ( θ t +1 ; R t +1 ) by definition of θ t +1 . We still need to show: L ( θ t ; R t +1 ) = max R ∈R n,k L ( θ t +1 ; R ) . We’ll give two proofs. 35 / 70

  46. Proof. We’ve already shown: ◮ L ( θ t ; R t +1 ) = L ( θ t ) ; ◮ L ( θ t ; R t +1 ) = max θ ∈ Θ L ( θ ; R t +1 ) ≤ L ( θ t +1 ; R t +1 ) by definition of θ t +1 . We still need to show: L ( θ t ; R t +1 ) = max R ∈R n,k L ( θ t +1 ; R ) . We’ll give two proofs. By concavity of ln (“Jensen’s inequality” in convexity lectures), for any R ∈ R n,k , n k R ij ln p θ t ( x i , y i = j ) � � L ( θ t ; R ) = R ij i =1 j =1   n k p θ t ( x i , y i = j ) � � ≤ ln R ij   R ij i =1 j =1 n � = ln p θ t ( x i ) = L ( θ t ) = L ( θ t ; R t +1 ) . i =1 Since R was arbitrary, max R ∈R L ( θ t ; R ) = L ( θ t ; R t +1 ) . 35 / 70

  47. Proof (continued). Here’s a second proof of that missing fact. To evaluate arg max R ∈R n,k L ( θ ; R ) , consider Lagrangian   n k k � k � � � � �  . R ij ln p θ ( x i , y = j ) − R ij ln R ij + λ i R ij − 1  i =1 j =1 j =1 j =1 Fixing i and taking the gradient with respect to R ij for any j , 0 = ln p θ ( x i , y i = j ) − ln R ij − 1 + λ i , giving R ij = p θ ( x i , y = j ) exp( λ i − 1) . Since moreover � � R ij = exp( λ i − 1) p θ ( x i , y = j ) = exp( λ i − 1) p θ ( x i ) , 1 = j j it follows that exp( λ i − 1) = 1 / p θ ( x i ) , and the optimal R satisfies R ij = p θ ( x i ,y = j ) / p θ ( x i ) = p θ ( y = j | x i ) . � 36 / 70

  48. Related issues. 37 / 70

  49. Parameter constraints. E-M for GMMs still works if we freeze or constrain some parameters. 38 / 70

  50. Parameter constraints. E-M for GMMs still works if we freeze or constrain some parameters. Examples: ◮ No weights: initialize π = ( 1 / k , . . . , 1 / k ) and never update it. ◮ Diagonal covariance matrices: update everything as before, except Σ j := diag(( σ j ) 2 1 , . . . , ( σ j ) 2 d ) where � n i =1 R ij ( x i − µ j ) 2 ( σ j ) 2 l l := ; nπ j that is: we use coordinate-wise sample variances weighted by R . Why is this a good idea? 38 / 70

  51. Parameter constraints. E-M for GMMs still works if we freeze or constrain some parameters. Examples: ◮ No weights: initialize π = ( 1 / k , . . . , 1 / k ) and never update it. ◮ Diagonal covariance matrices: update everything as before, except Σ j := diag(( σ j ) 2 1 , . . . , ( σ j ) 2 d ) where � n i =1 R ij ( x i − µ j ) 2 ( σ j ) 2 l l := ; nπ j that is: we use coordinate-wise sample variances weighted by R . Why is this a good idea? Computation (of inverse), sample complexity, . . . 38 / 70

  52. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  53. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  54. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  55. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  56. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  57. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  58. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  59. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  60. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  61. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  62. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  63. Gaussian Mixture Model with diagonal covariances. 10.0 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 5 0 5 10 15 39 / 70

  64. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  65. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  66. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  67. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  68. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  69. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  70. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  71. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  72. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  73. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

  74. Gaussian Mixture Model with diagonal covariances. 7.5 5.0 2.5 0.0 2.5 5.0 7.5 10.0 10 5 0 5 10 15 39 / 70

Recommend

Part 0: Git-ing Started Part 1: Essential Skills Part 2: Introduction to Git Part 3: Advanced

Part 0: Git-ing Started Part 1: Essential Skills Part 2: Introduction to Git Part 3: Advanced

Part 0: Git-ing Started Part 1: Essential Skills Part 2: Introduction to Git Part 3: Advanced Features Part 4: Shells Comic Credit: Randall Munroe, xkcd.com Lab 3: Version Control Part 0: Git-ing Started Part 1: Essential Skills Part 2:

514 views • 37 slides
Introduction Part One: Initial Problem Part Two: Progress Over Six Months Part

Introduction Part One: Initial Problem Part Two: Progress Over Six Months Part

Introduction Part One: Initial Problem Part Two: Progress Over Six Months Part Three: Key Outputs/Deliverables Part Four: What Have We Learned? Part Five: Next Steps Problem is Why spending entities dont execute their

751 views • 26 slides
Outline Part I. Introduction Part II. ML for DI Part III. DI for ML Training data

Outline Part I. Introduction Part II. ML for DI Part III. DI for ML Training data

Outline Part I. Introduction Part II. ML for DI Part III. DI for ML Training data creation Data cleaning Part IV. Conclusions and research directions Successful ML requires Data Integration Large collections of manually

510 views • 15 slides
DMR - Part 2 of 3 May 2, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3 -

DMR - Part 2 of 3 May 2, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3 -

R A R A Academy on DMR 05/02/2020 Rochester Amateur Radio Association, Inc. DMR - Part 2 of 3 May 2, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3 - Steve Verzulli KA1CNF http://rochesterham.org 1 Introduction

734 views • 20 slides
Commercial Dog Breeders Part 8: Housing (Part 2) Introduction Housing Part 1 Housing Part 2

Commercial Dog Breeders Part 8: Housing (Part 2) Introduction Housing Part 1 Housing Part 2

Introductory Course for Commercial Dog Breeders Part 8: Housing (Part 2) Introduction Housing Part 1 Housing Part 2 Specific requirements for Define types of facilities each type of facility: General requirements for Primary

1.12k views • 68 slides
Fusion - Part 3 of 3 May 16, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3

Fusion - Part 3 of 3 May 16, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3

R A R A Academy 05/16/2020 Rochester Amateur Radio Association, Inc. Fusion - Part 3 of 3 May 16, 2020 Part 1 - Mike Moore KC2NM Part 2 - Rich Hoffarth K2AXP Part 3 - Steve Verzulli KA1CNF http://rochesterham.org 1 Credits I wish thank

379 views • 19 slides
CS 309: Autonomous Intelligent Robotics FRI I Lecture 4: AI Part 2 & C++ Part 2 Instructor:

CS 309: Autonomous Intelligent Robotics FRI I Lecture 4: AI Part 2 & C++ Part 2 Instructor:

CS 309: Autonomous Intelligent Robotics FRI I Lecture 4: AI Part 2 & C++ Part 2 Instructor: Justin Hart http://justinhart.net/teaching/2019_spring_cs309/ Today What is Artificial Intelligence? Part 2 C++ Primer Part 2 Areas

423 views • 27 slides
01 INTR IN TRODUC UCTIO TION PART PE is a LIFE LI LIFE PA PAYT - To Tool to to re reduce

01 INTR IN TRODUC UCTIO TION PART PE is a LIFE LI LIFE PA PAYT - To Tool to to re reduce

01. PART INTRODUCTION AND MAIN OBJECTIVES Our LIFE2015/ENV/PT/609 agenda 02. PART METHODOLOGY 03. PART IMPLEMENTATION ADVISORY BOARD AND PROJECT BOARD MEETING 29-30 January 2018 04. PART CONCLUSION ABOU ABOUT T US 01 INTR IN

716 views • 17 slides
Fundraising 101 Fundraising Class Agenda Part I: Introductions Part II: Traditional Fundraising

Fundraising 101 Fundraising Class Agenda Part I: Introductions Part II: Traditional Fundraising

Fundraising 101 Fundraising Class Agenda Part I: Introductions Part II: Traditional Fundraising Practices Part III: Crowdfunding Basics Part IV: Grantseeking - 360 degree view Part V: Q&A 2 Why Fundraise? 1) Financing your creative practice

618 views • 31 slides
Wind Part 1: How do we measure it? Part 2: What exactly is wind? Part 3: Where is it? PART 1:

Wind Part 1: How do we measure it? Part 2: What exactly is wind? Part 3: Where is it? PART 1:

Wind Part 1: How do we measure it? Part 2: What exactly is wind? Part 3: Where is it? PART 1: Wind speed = Measuring anemometer Wind Wind direction = wind vane given from the direction it came FROM. Wind chill = increased

428 views • 20 slides
Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph

Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph

Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph Extraction Construction Part 4: Critical Analysis 1 Tutorial Outline 1. Knowledge Graph Primer [Jay] 2.

1.16k views • 66 slides
Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph

Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph

Part 1: Knowledge Graphs Part 2: Part 3: Knowledge Graph Extraction Construction Part 4: Critical Analysis 1 Tutorial Outline 1. Knowledge Graph Primer [Jay] 2.

869 views • 71 slides
The heartful PRESENTER Influence minds and win hearts Contents 04 PART 1 INTRODUCTION 06

The heartful PRESENTER Influence minds and win hearts Contents 04 PART 1 INTRODUCTION 06

The heartful PRESENTER Influence minds and win hearts Contents 04 PART 1 INTRODUCTION 06 PART 2 BODY LANGUAGE 08 PART 3 VOICE MODULATION 11 PART 4 POWER OF PAUSE 14 PART 5 PRESENCE 16 PART 6 EMPATHY THe

755 views • 17 slides
Project Presentation TAKING COOPERATION FORWARD 1 PROLINE-CE: THE PROJECT Part 1 Part 2 Part 3

Project Presentation TAKING COOPERATION FORWARD 1 PROLINE-CE: THE PROJECT Part 1 Part 2 Part 3

PROLINE-CE Project Presentation TAKING COOPERATION FORWARD 1 PROLINE-CE: THE PROJECT Part 1 Part 2 Part 3 Part 4 Funding Facts & Figures Objectives & Structure & Programme Timeline Partnership T arget groups Goals

429 views • 20 slides
Influence Identification on Independent Cascade Model

Influence Identification on Independent Cascade Model

Influence Identification on Independent Cascade Model Part 1 Introduction Part 2 Related Work Part 3 Algorithm Part 4 Experiment Part 5 Conclusion Part 6 References

723 views • 33 slides
Taking Zu as an Example Contents Introduction PART 1 Previous Studies PART 2 Data

Taking Zu as an Example Contents Introduction PART 1 Previous Studies PART 2 Data

Lexical Semantic Evolution of Polysemous Words in Ancient Chinese Taking Zu as an Example Contents Introduction PART 1 Previous Studies PART 2 Data Collection & Analysis PART 3 Conclusion & Future Work PART 4 2 Introduction

698 views • 20 slides
Industrial Robots Industrial Robots Control Control Part 2 Control Control Part 2 Part 2

Industrial Robots Industrial Robots Control Control Part 2 Control Control Part 2 Part 2

Industrial Robots Industrial Robots Control Control Part 2 Control Control Part 2 Part 2 Part 2 Introduction to centralized control Independent joint decentralized control may prove inadequate when the user requires high task

673 views • 36 slides
Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April

Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April

Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April 24 October 3 August 14 Linda Kara Reuter Melissa Hofschire Bowles-Terry User-centered Assessment: Digging into Take Action: Using

832 views • 43 slides
Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April

Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April

Webinar Series: Evaluating and Sharing Your Librarys Impact Part 1: Part 3: Part 2: April 24 October 3 August 14 Linda Kara Reuter Melissa Hofschire Bowles-Terry User-centered Assessment: Digging into Take Action: Using

1.11k views • 61 slides
10 GHz (Microwave), up North Part #1 Part #2 Part #3 Promotion Activities Planning Parts 2

10 GHz (Microwave), up North Part #1 Part #2 Part #3 Promotion Activities Planning Parts 2

10 GHz (Microwave), up North Part #1 Part #2 Part #3 Promotion Activities Planning Parts 2 & 3 Presented at the NTMS Microwave Mini-conference on Nov. 9 th 2013 by Jim Froemke K0MHC/rover 1 10 GHz DXpedition to the Great Lakes 3 4

693 views • 45 slides
CS188 Outline Were done with Part I: Search and Planning! Part II: Probabilistic

CS188 Outline Were done with Part I: Search and Planning! Part II: Probabilistic

CS188 Outline Were done with Part I: Search and Planning! Part II: Probabilistic Reasoning Diagnosis Speech recognition Tracking objects Robot mapping Genetics Error correcting codes lots more! Part

530 views • 18 slides
PROCESS SKID PACKAGES INDEX Part I : The Company, Quality System & Certification Part

PROCESS SKID PACKAGES INDEX Part I : The Company, Quality System & Certification Part

PROCESS SKID PACKAGES INDEX Part I : The Company, Quality System & Certification Part II : Euroguarco Manufacturing & Supply Part III : Engineering Activities Part IV: Pressure Equipment Part V: Instrumentation Packages

560 views • 40 slides
FY17 CONSOLIDATED RESULTS UNIPOL AND UNIPOLSAI Bologna, 23 March 2018 2 PART 1 PART 2 PART 3

FY17 CONSOLIDATED RESULTS UNIPOL AND UNIPOLSAI Bologna, 23 March 2018 2 PART 1 PART 2 PART 3

FY17 CONSOLIDATED RESULTS UNIPOL AND UNIPOLSAI Bologna, 23 March 2018 2 PART 1 PART 2 PART 3 Consolidated Insurance Banking Business results Business Non-Life Life PART 4 PART 5 PART 6 Investments Dividend proposal

430 views • 30 slides
Outline Overview Part I: Emacs ESS tutorial, UseR! 2011 meeting part II: ESS Stephen Eglen

Outline Overview Part I: Emacs ESS tutorial, UseR! 2011 meeting part II: ESS Stephen Eglen

Outline Overview Part I: Emacs ESS tutorial, UseR! 2011 meeting part II: ESS Stephen Eglen Part III: Sweave Part IV: Org mode 2011-08-15 Advanced topics Part V Future steps / Q & A session / open discussion End Aims of the tutorial

847 views • 26 slides

More recommend