Probabilistic Inference Lecture 1 M. Pawan Kumar pawan.kumar@ecp.fr Slides available online http://cvc.centrale-ponts.fr/personnel/pawan/
About the Course 7 lectures + 1 exam Probabilistic Models – 1 lecture Energy Minimization – 4 lectures Computing Marginals – 2 lectures Related Courses Probabilistic Graphical Models (MVA) Structured Prediction
Instructor Assistant Professor (2012 – Present) Center for Visual Computing 12 Full-time Faculty Members 2 Associate Faculty Members Research Interests Probabilistic Models Machine Learning Computer Vision Medical Image Analysis
Students Third year at ECP Specializing in Machine Learning and Vision Prerequisites Probability Theory Continuous Optimization Discrete Optimization
Outline Probabilistic Models Conversions Exponential Family Inference Example (on board) !!
Outline Probabilistic Models Markov Random Fields (MRF) Bayesian Networks Factor Graphs Conversions Exponential Family Inference
MRF Unobserved Random Variables Neighbors Edges define a neighborhood over random variables
MRF Variable Va takes a value or a label va from a set L = {l1, l2,…, lh} V = v is called a labeling Discrete, Finite
MRF MRF assumes the Markovian property for P(v) V1 V2 V3 V4 V5 V6 V7
MRF Va is conditionally independent of Vb given Va’s neighbors Hammersley-Clifford Theorem
MRF Potential ψ12(v1,v2) Potential ψ56(v5,v6) Probability P(v) can be decomposed into clique potentials
MRF Potential ψ1(v1,d1) Observed Data Probability P(v) proportional to Π(a,b) ψab(va,vb) Probability P(d|v) proportional to Πa ψa (va,da)
MRF Πa ψa(va,da) Π(a,b) ψab(va,vb) Probability P(v,d) = Z Z is known as the partition function
MRF High-order Potential ψ4578(v4,v5,v7,v8) d1 d2 d3 V1 V2 V3 d4 d5 d6
Pairwise MRF Unary Potential ψ1(v1,d1) Pairwise Potential ψ56(v5,v6) Πa ψa(va,da) Π(a,b) ψab(va,vb) Probability P(v,d) = Z Z is known as the partition function
MRF A is conditionally independent of B given C if there is no path from A to B when C is removed
Conditional Random Fields (CRF) V1 V2 V3 d4 d5 d6 V4 V5 V6 d7 d8 d9 V7 V8 V9 CRF assumes the Markovian property for P(v|d) Hammersley-Clifford Theorem
CRF Probability P(v|d) proportional to Πa ψa(va;d) Π(a,b) ψab(va,vb;d) Clique potentials that depend on the data
CRF Πa ψa (va;d) Π(a,b) ψab(va,vb;d) Probability P(v|d) = Z Z is known as the partition function
MRF and CRF Πa ψa(va) Π(a,b) ψab(va,vb) Probability P(v) = Z V1 V2 V3
Outline Probabilistic Models Markov Random Fields (MRF) Bayesian Networks Factor Graphs Conversions Exponential Family Inference
Bayesian Networks Directed Acyclic Graph (DAG) – no directed loops V1 V2 V3 V4 V5 V6 V7 V8 Directed Acyclic Graph (DAG) – no directed loops Ignoring directionality of edges, a DAG can have loops
Bayesian Networks V1 V2 V3 V4 V5 V6 V7 V8 Bayesian Network concisely represents the probability P(v)
Bayesian Networks Probability P(v) = Πa P(va|Parents(va)) P(v1)P(v2|v1)P(v3|v1)P(v4|v2)P(v5|v2,v3)P(v6|v3)P(v7|v4,v5)P(v8|v5,v6)
Bayesian Networks Courtesy Kevin Murphy
Bayesian Networks V1 V2 V3 V4 V5 V6 V7 V8 Va is conditionally independent of its ancestors given its parents
Bayesian Networks Conditional independence of A and B given C Courtesy Kevin Murphy
Outline Probabilistic Models Markov Random Fields (MRF) Bayesian Networks Factor Graphs Conversions Exponential Family Inference
Factor Graphs Two types of nodes: variable nodes and factor nodes Bipartite graph between the two types of nodes
Factor Graphs ψa(v1,v2) V1 V2 V3 a b c d e V4 V5 V6 f g Factor graphs concisely represents the probability P(v)
Factor Graphs ψa({v}a) b c d e V4 V5 V6 f g Factor graphs concisely represents the probability P(v)
Factor Graphs ψb(v2,v3) V1 V2 V3 a b c d e V4 V5 V6 f g Factor graphs concisely represents the probability P(v)
Factor Graphs ψb({v}b) d e V4 V5 V6 f g Factor graphs concisely represents the probability P(v)
Factor Graphs ψb({v}b) Πa ψa({v}a) Probability P(v) = Z d e V4 V5 V6 f g Πa ψa({v}a) Probability P(v) = Z Z is known as the partition function
Outline Probabilistic Models Conversions Exponential Family Inference
MRF to Factor Graphs
Bayesian Networks to Factor Graphs
Factor Graphs to MRF
Outline Probabilistic Models Conversions Exponential Family Inference
Motivation Random Variable V Label set L = {l1, l2,…, lh} Samples V1, V2, …, Vm that are i.i.d. Functions ϕα: L Reals α indexes a set of functions Empirical expectations: μα = (Σi ϕα(Vi))/m Expectation wrt distribution P: EP[ϕα(V)] = Σi ϕα(li)P(li) Given empirical expectations, find compatible distribution Underdetermined problem
Maximum Entropy Principle max Entropy of the distribution s.t. Distribution is compatible
Maximum Entropy Principle max -Σi P(li)log(P(li)) s.t. Distribution is compatible
Maximum Entropy Principle max -Σi P(li)log(P(li)) s.t. Σi ϕα(li)P(li) = μα for all α Σi P(li) = 1 P(v) proportional to exp(-Σα θαϕα(v))
Exponential Family Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2,…, lh} Labeling V = v, va L for all a {1, 2,…, n} Functions ϕα: Ln Reals α indexes a set of functions P(v) = exp{-Σα θαΦα(v) - A(θ)} Parameters Sufficient Statistics Normalization Constant
Minimal Representation P(v) = exp{-Σα θαΦα(v) - A(θ)} Parameters Sufficient Statistics Normalization Constant No non-zero c such that Σα cαΦα(v) = Constant
Ising Model P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2}
Ising Model P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {-1, +1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters va θa for all Va V vavb θab for all (Va,Vb) E
Ising Model P(v) = exp{-Σa θava -Σa,b θabvavb- A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {-1, +1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters va θa for all Va V vavb θab for all (Va,Vb) E
Interactive Binary Segmentation
Interactive Binary Segmentation Foreground histogram of RGB values FG Background histogram of RGB values BG ‘+1’ indicates foreground and ‘-1’ indicates background
Interactive Binary Segmentation More likely to be foreground than background
Interactive Binary Segmentation θa proportional to -log(FG(da)) + log(BG(da)) More likely to be background than foreground
Interactive Binary Segmentation More likely to belong to same label
Interactive Binary Segmentation θab proportional to -exp(-(da-db)2) Less likely to belong to same label
Rest of lecture 1 ….
Exponential Family P(v) = exp{-Σα θαΦα(v) - A(θ)} Parameters Sufficient Statistics Log-Partition Function Random Variables V = {V1,V2,…,Vn} Random Variable Va takes a value or label va va L = {l1,l2,…,lh} Labeling V = v
Overcomplete Representation P(v) = exp{-Σα θαΦα(v) - A(θ)} Parameters Sufficient Statistics Log-Partition Function There exists a non-zero c such that Σα cαΦα(v) = Constant
Ising Model P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2}
Ising Model P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {0, 1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters Ia;i(va) θa;i for all Va V, li L θab;ik Iab;ik(va,vb) for all (Va,Vb) E, li, lk L Ia;i(va): indicator for va = li Iab;ik(va,vb): indicator for va = li, vb = lk
Ising Model P(v) = exp{-Σa Σi θa;iIa;i(va) -Σa,b Σi,k θab;ikIab;ik(va,vb) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {0, 1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters Ia;i(va) θa;i for all Va V, li L θab;ik Iab;ik(va,vb) for all (Va,Vb) E, li, lk L Ia;i(va): indicator for va = li Iab;ik(va,vb): indicator for va = li, vb = lk
Interactive Binary Segmentation Foreground histogram of RGB values FG Background histogram of RGB values BG ‘1’ indicates foreground and ‘0’ indicates background
Interactive Binary Segmentation More likely to be foreground than background
Interactive Binary Segmentation θa;0 proportional to -log(BG(da)) θa;1 proportional to -log(FG(da)) More likely to be background than foreground
Interactive Binary Segmentation More likely to belong to same label
Interactive Binary Segmentation θab;ik proportional to exp(-(da-db)2) if i ≠ k θab;ik = 0 if i = k Less likely to belong to same label
Metric Labeling P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh}
Metric Labeling P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {0, …, h-1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters Ia;i(va) θa;i for all Va V, li L θab;ik Iab;ik(va,vb) for all (Va,Vb) E, li, lk L θab;ik is a metric distance function over labels
Metric Labeling P(v) = exp{-Σa Σi θa;iIa;i(va) -Σa,b Σi,k θab;ikIab;ik(va,vb) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {0, …, h-1} Neighborhood over variables specified by edges E Sufficient Statistics Parameters Ia;i(va) θa;i for all Va V, li L θab;ik Iab;ik(va,vb) for all (Va,Vb) E, li, lk L θab;ik is a metric distance function over labels
Stereo Correspondence Disparity Map
Stereo Correspondence L = {disparities} Pixel (xa,ya) in left corresponds to pixel (xa+va,ya) in right
Stereo Correspondence L = {disparities} θa;i is proportional to the difference in RGB values
Stereo Correspondence L = {disparities} θab;ik = wab d(i,k) wab proportional to exp(-(da-db)2)
Pairwise MRF P(v) = exp{-Σα θαΦα(v) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh} Neighborhood over variables specified by edges E Sufficient Statistics Parameters Ia;i(va) θa;i for all Va V, li L θab;ik Iab;ik(va,vb) for all (Va,Vb) E, li, lk L
Pairwise MRF P(v) = exp{-Σa Σi θa;iIa;i(va) -Σa,b Σi,k θab;ikIab;ik(va,vb) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh} Neighborhood over variables specified by edges E Πa ψa(va) Π(a,b) ψab(va,vb) Probability P(v) = Z A(θ) : log Z ψa(li) : exp(-θa;i) ψa(li,lk) : exp(-θab;ik) Parameters θ are sometimes also referred to as potentials
Pairwise MRF P(v) = exp{-Σa Σi θa;iIa;i(va) -Σa,b Σi,k θab;ikIab;ik(va,vb) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh} Neighborhood over variables specified by edges E Labeling as a function f : {1, 2, … , n} {1, 2, …, h} Variable Va takes a label lf(a)
Pairwise MRF P(f) = exp{-Σa θa;f(a) -Σa,b θab;f(a)f(b) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh} Neighborhood over variables specified by edges E Labeling as a function f : {1, 2, … , n} {1, 2, …, h} Variable Va takes a label lf(a) Energy Q(f) = Σa θa;f(a) + Σa,b θab;f(a)f(b)
Pairwise MRF P(f) = exp{-Q(f) - A(θ)} Random Variable V = {V1, V2, …,Vn} Label set L = {l1, l2, …, lh} Neighborhood over variables specified by edges E Labeling as a function f : {1, 2, … , n} {1, 2, …, h} Variable Va takes a label lf(a) Energy Q(f) = Σa θa;f(a) + Σa,b θab;f(a)f(b)
Outline Probabilistic Models Conversions Exponential Family Inference
Inference maxv ( P(v) = exp{-Σa Σi θa;iIa;i(va) -Σa,b Σi,k θab;ikIab;ik(va,vb) - A(θ)} ) Maximum a Posteriori (MAP) Estimation minf ( Q(f) = Σa θa;f(a) + Σa,b θab;f(a)f(b) ) Energy Minimization P(va = li) = Σv P(v)δ(va = li) P(va = li, vb = lk) = Σv P(v)δ(va = li)δ(vb = lk) Computing Marginals
Next Lecture … Energy minimization for tree-structured pairwise MRF