hilbert s nullstellensatz linear algebra and
play

Hilberts Nullstellensatz, Linear Algebra and Combinatorial Problems - PowerPoint PPT Presentation

Mar 24, 2024 •240 likes •1.8k views

Hilberts Nullstellensatz, Linear Algebra and Combinatorial Problems Susan Margulies Mathematics Department, US Naval Academy, Annapolis, Maryland Visiting National Institute of Standards and Technology! April 7, 2015 Susan Margulies, US

  1. Hilbert’s Nullstellensatz Theorem (1893): Let K be an algebraically closed field and f 1 , . . . , f s be polynomials in K [ x 1 , . . . , x n ]. Given a system of equations such that f 1 = f 2 = · · · = f s = 0 , then this system has no solution if and only if there exist polynomials β 1 , . . . , β s ∈ K [ x 1 , . . . , x n ] such that s � 1 = β i f i . ✷ i =1 � �� � This polynomial identity is a Nullstellensatz certificate . � � Definition: Let d = max deg( β 1 ) , deg( β 2 ) , . . . , deg( β s ) . Then d is the degree of the Nullstellensatz certificate . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  2. Nullstellensatz Degree Upper Bounds Recall n is the number of variables, and the number of monomials � n + d − 1 � of degree d in n variables is . n − 1 Theorem: (Koll´ ar, 1988) The deg( β i ) is bounded by � �� n � deg( β i ) ≤ max 3 , max { deg( f i ) } . (bound is tight for certain pathologically bad examples) Theorem: (Lazard 1977, Brownawell 1987) The deg( β i ) is bounded by � � deg( β i ) ≤ n max { deg( f i ) } − 1 . (bound applies to particular zero-dimensional ideals) Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  3. Nullstellensatz Degree Upper Bounds Recall n is the number of variables, and the number of monomials � n + d − 1 � of degree d in n variables is . n − 1 Theorem: (Koll´ ar, 1988) The deg( β i ) is bounded by � �� n � deg( β i ) ≤ max 3 , max { deg( f i ) } . (bound is tight for certain pathologically bad examples) Theorem: (Lazard 1977, Brownawell 1987) The deg( β i ) is bounded by � � deg( β i ) ≤ n max { deg( f i ) } − 1 . (bound applies to particular zero-dimensional ideals) Question: What about lower bounds? How do we find them? Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  4. N ul LA running on a particular instance: INPUT: A system of polynomial equations x 2 1 − 1 = 0 , x 1 + x 3 = 0 , x 1 + x 2 = 0 , x 2 + x 3 = 0 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  5. N ul LA running on a particular instance: INPUT: A system of polynomial equations x 2 1 − 1 = 0 , x 1 + x 3 = 0 , x 1 + x 2 = 0 , x 2 + x 3 = 0 1 Construct a hypothetical Nullstellensatz certificate of degree 1 ( x 2 1 = ( c 0 x 1 + c 1 x 2 + c 2 x 3 + c 3 ) 1 − 1) + ( c 4 x 1 + c 5 x 2 + c 6 x 3 + c 7 ) ( x 1 + x 2 ) � �� � � �� � β 1 β 2 + ( c 8 x 1 + c 9 x 2 + c 10 x 3 + c 11 ) ( x 1 + x 3 ) + ( c 12 x 1 + c 13 x 2 + c 14 x 3 + c 15 ) ( x 2 + x 3 ) � �� � � �� � β 3 β 4 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  6. N ul LA running on a particular instance: INPUT: A system of polynomial equations x 2 1 − 1 = 0 , x 1 + x 3 = 0 , x 1 + x 2 = 0 , x 2 + x 3 = 0 1 Construct a hypothetical Nullstellensatz certificate of degree 1 ( x 2 1 = ( c 0 x 1 + c 1 x 2 + c 2 x 3 + c 3 ) 1 − 1) + ( c 4 x 1 + c 5 x 2 + c 6 x 3 + c 7 ) ( x 1 + x 2 ) � �� � � �� � β 1 β 2 + ( c 8 x 1 + c 9 x 2 + c 10 x 3 + c 11 ) ( x 1 + x 3 ) + ( c 12 x 1 + c 13 x 2 + c 14 x 3 + c 15 ) ( x 2 + x 3 ) � �� � � �� � β 3 β 4 2 Expand the hypothetical Nullstellensatz certificate c 0 x 3 1 + c 1 x 2 1 x 2 + c 2 x 2 1 x 3 + ( c 3 + c 4 + c 8 ) x 2 1 + ( c 5 + c 13 ) x 2 2 + ( c 10 + c 14 ) x 2 3 + ( c 4 + c 5 + c 9 + c 12 ) x 1 x 2 + ( c 6 + c 8 + c 10 + c 12 ) x 1 x 3 + ( c 6 + c 9 + c 13 + c 14 ) x 2 x 3 + ( c 7 + c 11 − c 0 ) x 1 + ( c 7 + c 15 − c 1 ) x 2 + ( c 11 + c 15 − c 2 ) x 3 − c 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  7. N ul LA running on a particular instance: INPUT: A system of polynomial equations x 2 1 − 1 = 0 , x 1 + x 3 = 0 , x 1 + x 2 = 0 , x 2 + x 3 = 0 1 Construct a hypothetical Nullstellensatz certificate of degree 1 ( x 2 1 = ( c 0 x 1 + c 1 x 2 + c 2 x 3 + c 3 ) 1 − 1) + ( c 4 x 1 + c 5 x 2 + c 6 x 3 + c 7 ) ( x 1 + x 2 ) � �� � � �� � β 1 β 2 + ( c 8 x 1 + c 9 x 2 + c 10 x 3 + c 11 ) ( x 1 + x 3 ) + ( c 12 x 1 + c 13 x 2 + c 14 x 3 + c 15 ) ( x 2 + x 3 ) � �� � � �� � β 3 β 4 2 Expand the hypothetical Nullstellensatz certificate c 0 x 3 1 + c 1 x 2 1 x 2 + c 2 x 2 1 x 3 + ( c 3 + c 4 + c 8 ) x 2 1 + ( c 5 + c 13 ) x 2 2 + ( c 10 + c 14 ) x 2 3 + ( c 4 + c 5 + c 9 + c 12 ) x 1 x 2 + ( c 6 + c 8 + c 10 + c 12 ) x 1 x 3 + ( c 6 + c 9 + c 13 + c 14 ) x 2 x 3 + ( c 7 + c 11 − c 0 ) x 1 + ( c 7 + c 15 − c 1 ) x 2 + ( c 11 + c 15 − c 2 ) x 3 − c 3 3 Extract a linear system of equations from expanded certificate c 0 = 0 , . . . , c 3 + c 4 + c 8 = 0 , c 11 + c 15 − c 2 = 0 , − c 3 = 1 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  8. N ul LA running on a particular instance: c 0 c 1 c 2 c 3 c 4 c 5 c 6 c 7 c 8 c 9 c 10 c 11 c 12 c 13 c 14 c 15 x 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 x 2 1 x 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 x 2 1 x 3 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 x 2 0 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0 1 x 2 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 2 x 2 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 3 0 0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 0 x 1 x 2 0 0 0 0 0 0 1 0 1 0 1 0 1 0 0 0 0 x 1 x 3 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 0 x 2 x 3 − 1 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 x 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 x 2 0 0 − 1 0 0 0 0 0 0 0 0 1 0 0 0 1 0 x 3 1 0 0 0 − 1 0 0 0 0 0 0 0 0 0 0 0 0 1 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  9. N ul LA running on a particular instance: 4 Solve the linear system, and assemble the certificate 1 − 1) + 1 2 x 1 ( x 1 + x 2 ) − 1 2 x 1 ( x 2 + x 3 ) + 1 1 = − ( x 2 2 x 1 ( x 1 + x 3 ) Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  10. N ul LA running on a particular instance: 4 Solve the linear system, and assemble the certificate 1 − 1) + 1 2 x 1 ( x 1 + x 2 ) − 1 2 x 1 ( x 2 + x 3 ) + 1 1 = − ( x 2 2 x 1 ( x 1 + x 3 ) 5 Otherwise, increment the degree and repeat. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  11. N ul LA Summary INPUT: A system of polynomial equations Construct a hypothetical Nullstellensatz certificate of degree d 1 Expand the hypothetical Nullstellensatz certificate 2 Extract a linear system of equations from expanded certificate 3 Solve the linear system. 4 If there is a solution, assemble the certificate. 1 Otherwise, loop and repeat with a larger degree d until known 2 upper bounds are exceeded. OUTPUT: yes , there is a solution. 1 no , there is no solution, along with a certificate of infeasibility . 2 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  12. Partition Problem: Definition and Example Partition: Given set of integers W = { w 1 , . . . , w n } , can W be partitioned into two sets, S and W \ S such that � � w = w . w ∈ S w ∈ W \ S Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  13. Partition Problem: Definition and Example Partition: Given set of integers W = { w 1 , . . . , w n } , can W be partitioned into two sets, S and W \ S such that � � w = w . w ∈ S w ∈ W \ S Example: Let W = { 1 , 3 , 5 , 7 , 7 , 9 } . Then � �� � ���� S W \ S . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  14. Partition Problem: Definition and Example Partition: Given set of integers W = { w 1 , . . . , w n } , can W be partitioned into two sets, S and W \ S such that � � w = w . w ∈ S w ∈ W \ S Example: Let W = { 1 , 3 , 5 , 7 , 7 , 9 } . Then � �� � ���� S W \ S 1 + 3 + 5 + 7 . � �� � S Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  15. Partition Problem: Definition and Example Partition: Given set of integers W = { w 1 , . . . , w n } , can W be partitioned into two sets, S and W \ S such that � � w = w . w ∈ S w ∈ W \ S Example: Let W = { 1 , 3 , 5 , 7 , 7 , 9 } . Then � �� � ���� S W \ S 1 + 3 + 5 + 7 7 + 9 . � �� � � �� � S W \ S Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  16. Partition Problem: Definition and Example Partition: Given set of integers W = { w 1 , . . . , w n } , can W be partitioned into two sets, S and W \ S such that � � w = w . w ∈ S w ∈ W \ S Example: Let W = { 1 , 3 , 5 , 7 , 7 , 9 } . Then � �� � ���� S W \ S 16 = 1 + 3 + 5 + 7 = 7 + 9 = 16 . � �� � � �� � S W \ S Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  17. Partition as a System of Polynomial Equations Given a set of integers W = { w 1 , . . . , w n } : one variable per integer : x 1 , . . . , x n Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  18. Partition as a System of Polynomial Equations Given a set of integers W = { w 1 , . . . , w n } : one variable per integer : x 1 , . . . , x n For i = 1 , . . . , n , let x 2 i − 1 = 0 . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  19. Partition as a System of Polynomial Equations Given a set of integers W = { w 1 , . . . , w n } : one variable per integer : x 1 , . . . , x n For i = 1 , . . . , n , let x 2 i − 1 = 0 . and finally, n � w i x i = 0 . i =1 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  20. Partition as a System of Polynomial Equations Given a set of integers W = { w 1 , . . . , w n } : one variable per integer : x 1 , . . . , x n For i = 1 , . . . , n , let x 2 i − 1 = 0 . and finally, n � w i x i = 0 . i =1 Proposition: Given a set of integers W = { w 1 , . . . , w n } , the above system of n + 1 polynomial equations has a solution if and only if there exists a partition of W into two sets, S ⊆ W and W \ S , such that � w ∈ S w = � w ∈ W \ S w . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  21. Partition as a System of Polynomial Equations Given a set of integers W = { w 1 , . . . , w n } : one variable per integer : x 1 , . . . , x n For i = 1 , . . . , n , let x 2 i − 1 = 0 . and finally, n � w i x i = 0 . i =1 Proposition: Given a set of integers W = { w 1 , . . . , w n } , the above system of n + 1 polynomial equations has a solution if and only if there exists a partition of W into two sets, S ⊆ W and W \ S , such that � w ∈ S w = � w ∈ W \ S w . Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  22. NP, coNP and the Nullstellensatz Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  23. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  24. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. Definition coNP is the class of problems whose complements are in NP . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  25. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. Definition coNP is the class of problems whose complements are in NP . It is widely believed that coNP � = NP . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  26. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. (hard to find) Definition coNP is the class of problems whose complements are in NP . It is widely believed that coNP � = NP . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  27. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. (hard to find) Definition coNP is the class of problems whose complements are in NP . (hard to verify) It is widely believed that coNP � = NP . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  28. NP, coNP and the Nullstellensatz Definition NP is the class of problems whose solutions can be verified in polynomial-time. (hard to find) Definition coNP is the class of problems whose complements are in NP . (hard to verify) Observation The Partition problem is NP-complete. It is widely believed that coNP � = NP . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  29. Minimum-degree Nullstellensatz Certificates Example Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  30. Minimum-degree Nullstellensatz Certificates Example Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? Answer: No! x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 1 − 1 ) 3465 x 3 x 4 � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 1155 x 3 x 4 2 − 1 ) � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 3465 x 2 x 3 4 − 1 ) � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 3465 x 2 x 3 x 4 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  31. Minimum-degree Partition Nullstellensatz Certificates Let S n k denote the set of k -subsets of { 1 , . . . , n } Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  32. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  33. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  34. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  35. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  36. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  37. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  38. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  39. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  40. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  41. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: degree is n for n odd and n − 1 for n even. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  42. Minimum-degree Partition Nullstellensatz Certificates � �� � n Let S n i.e., | S n k denote the set of k -subsets of { 1 , . . . , n } k | = k Theorem (S.M., S. Onn, 2012) Given a set of non-partitionable integers W = { w 1 , . . . , w n } encoded as a system of polynomial equations as above, there exists a minimum-degree Nullstellensatz certificate for the non-existence of a partition of W as follows: n n � � c i , s x s � � � b s x s �� � � � � � ( x 2 1 = i − 1) + w i x i . s ∈ S n i =1 s ∈ S n \ i i =1 k even k odd k k ≤ n − 1 k ≤ n k Moreover, every Nullstellensatz certificate associated with the above system of polynomial equations contains exactly one monomial for each of the even parity subsets of S n \ i , and exactly k one monomial for each of the odd parity subsets of S n k . Note: certificate is both high degree and dense. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  43. Minimum-degree Nullstellensatz Certificates Example Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  44. Minimum-degree Nullstellensatz Certificates Example Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? Answer: No! x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 1 − 1 ) 3465 x 3 x 4 � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 1155 x 3 x 4 2 − 1 ) � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 3465 x 2 x 3 4 − 1 ) � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 3465 x 2 x 3 x 4 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  45. Minimum-degree Nullstellensatz Certificates Example Question: Let W = { 1 , 3 , 5 , 2 } . Is W partitionable? Answer: No! x 2 x 2 x 3 x 2 1 − 1 = 0 , 2 − 1 = 0 , 3 − 1 = 0 , 4 − 1 = 0 , x 1 + 3 x 2 + 5 x 3 + 2 x 4 = 0 . � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 1 − 1 ) 3465 x 3 x 4 � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 1155 x 3 x 4 2 − 1 ) � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 3465 x 2 x 3 4 − 1 ) � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 3465 x 2 x 3 x 4 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  46. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } .   0 w 3 w 2 w 1 w 2 w 3 0 w 1     0 w 1 w 3 w 2   0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  47. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . w 3   0 w 3 w 2 w 1 w 3 w 2 w 3 0 w 1   w 3   0 w 1 w 3 w 2   w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  48. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . w 1 w 2 w 3   0 w 3 w 2 w 1 w 3 w 2 w 3 0 w 1   w 3   0 w 1 w 3 w 2   w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  49. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . w 1 w 2 w 3   0 w 3 w 2 w 1 w 1 w 2 w 3 w 2 w 3 0 w 1   w 3   0 w 1 w 3 w 2   w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  50. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . w 1 w 2 w 3   0 w 3 w 2 w 1 w 1 w 2 w 3 w 2 w 3 0 w 1   w 2 w 1 w 3   0 w 1 w 3 w 2   w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  51. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . w 1 w 2 w 3   0 w 3 w 2 w 1 w 1 w 2 w 3 w 2 w 3 0 w 1   w 2 w 1 w 3   0 w 1 w 3 w 2   w 1 w 2 w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  52. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . − + w 1 + w 2 + w 3   0 w 3 w 2 w 1 − w 1 + w 2 + w 3 w 2 w 3 0 w 1   − w 2 + w 1 + w 3   0 w 1 w 3 w 2   − w 1 − w 2 + w 3 0 w 1 w 2 w 3 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  53. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . − + w 1 + w 2 + w 3   0 w 3 w 2 w 1 − w 1 + w 2 + w 3 w 2 w 3 0 w 1   − w 2 + w 1 + w 3   0 w 1 w 3 w 2   − w 1 − w 2 + w 3 0 w 1 w 2 w 3 ( w 1 + w 2 + w 3 )( − w 1 + w 2 + w 3 )( w 1 − w 2 + w 3 )( − w 1 − w 2 + w 3 ) � �� � partition polynomial Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  54. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . − + w 1 + w 2 + w 3   0 w 3 w 2 w 1 − w 1 + w 2 + w 3 w 2 w 3 0 w 1   − w 2 + w 1 + w 3   0 w 1 w 3 w 2   − w 1 − w 2 + w 3 0 w 1 w 2 w 3 ( w 1 + w 2 + w 3 )( − w 1 + w 2 + w 3 )( w 1 − w 2 + w 3 )( − w 1 − w 2 + w 3 ) � �� � partition polynomial Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  55. The Partition Matrix: Extract a Square Linear System Let W = { w 1 , w 2 , w 3 } . − + w 1 + w 2 + w 3   0 w 3 w 2 w 1 − w 1 + w 2 + w 3 w 2 w 3 0 w 1   − w 2 + w 1 + w 3   0 w 1 w 3 w 2   − w 1 − w 2 + w 3 0 w 1 w 2 w 3 The determinant of the above partition matrix is the ( w 1 + w 2 + w 3 )( − w 1 + w 2 + w 3 )( w 1 − w 2 + w 3 )( − w 1 − w 2 + w 3 ) � �� � partition polynomial Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  56. Another Example of the Partition Matrix Let W = { w 1 , . . . , w 4 } . The partition matrix P is   w 4 w 3 w 2 w 1 0 0 0 0 0 0 0 0 w 3 w 4 w 2 w 1     w 2 0 w 4 0 w 3 0 w 1 0     w 1 0 0 w 4 0 w 3 w 2 0   P = ,   0 w 2 w 3 0 w 4 0 0 w 1     0 w 1 0 w 3 0 w 4 0 w 2     0 0 0 0  w 1 w 2 w 4 w 3  0 0 0 0 w 1 w 2 w 3 w 4 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  57. Another Example of the Partition Matrix Let W = { w 1 , . . . , w 4 } . The partition matrix P is   w 4 w 3 w 2 w 1 0 0 0 0 0 0 0 0 w 3 w 4 w 2 w 1     w 2 0 w 4 0 w 3 0 w 1 0     w 1 0 0 w 4 0 w 3 w 2 0   P = ,   0 w 2 w 3 0 w 4 0 0 w 1     0 w 1 0 w 3 0 w 4 0 w 2     0 0 0 0  w 1 w 2 w 4 w 3  0 0 0 0 w 1 w 2 w 3 w 4 det( P ) = ( w 1 + w 2 + w 3 + w 4 )( − w 1 + w 2 + w 3 + w 4 )( w 1 − w 2 + w 3 + w 4 ) ( w 1 + w 2 − w 3 + w 4 )( − w 1 + w 2 − w 3 + w 4 )( − w 1 − w 2 + w 3 + w 4 ) ( w 1 − w 2 − w 3 + w 4 )( − w 1 − w 2 − w 3 + w 4 ) . “partition polynomial” Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  58. Determinant and Partition Polynomial Theorem (S.M., S. Onn, 2012) The determinant of the partition matrix is the partition polynomial. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  59. Hilbert’s Nullstellensatz Numeric Coefficients and the Partition Polynomial Given a square non-singular matrix A , Cramer’s rule states that Ax = b can be solved according to the formula x i = det( A | i b ) , det( A ) where A | i b is the matrix A with the i -th column replaced with the right-hand side vector b . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  60. Recall the non-partitionable W = { 1 , 3 , 5 , 2 } : � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 3465 x 3 x 4 1 − 1 ) � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 2 − 1 ) 1155 x 3 x 4 � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 4 − 1 ) 3465 x 2 x 3 � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . 3465 x 2 x 3 x 4 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  61. Recall the non-partitionable W = { 1 , 3 , 5 , 2 } : � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 3465 x 3 x 4 1 − 1 ) � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 2 − 1 ) 1155 x 3 x 4 � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 4 − 1 ) 3465 x 2 x 3 � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . 3465 x 2 x 3 x 4 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  62. Recall the non-partitionable W = { 1 , 3 , 5 , 2 } : � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 3465 x 3 x 4 1 − 1 ) � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 2 − 1 ) 1155 x 3 x 4 � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 4 − 1 ) 3465 x 2 x 3 � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . 3465 x 2 x 3 x 4 − 51975 = (1 + 3 + 5 + 2)( − 1 + 3 + 5 + 2)(1 − 3 + 5 + 2)(1 + 3 − 5 + 2) ( − 1 − 3 + 5 + 2)( − 1 + 3 − 5 + 2)(1 − 3 − 5 + 2)( − 1 − 3 − 5 + 2) . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  63. Recall the non-partitionable W = { 1 , 3 , 5 , 2 } : � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 3465 x 3 x 4 1 − 1 ) � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 2 − 1 ) 1155 x 3 x 4 � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 4 − 1 ) 3465 x 2 x 3 � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . 3465 x 2 x 3 x 4 − 51975 = (1 + 3 + 5 + 2)( − 1 + 3 + 5 + 2)(1 − 3 + 5 + 2)(1 + 3 − 5 + 2) ( − 1 − 3 + 5 + 2)( − 1 + 3 − 5 + 2)(1 − 3 − 5 + 2)( − 1 − 3 − 5 + 2) . Via Cramer’s rule, we see that the unknown b 4 is equal to b 4 = − 2550 − 51975 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  64. Recall the non-partitionable W = { 1 , 3 , 5 , 2 } : � � − 155 693 + 842 3465 x 2 x 3 − 188 693 x 2 x 4 + 908 ( x 2 1 = 3465 x 3 x 4 1 − 1 ) � � 231 + 842 1 1155 x 1 x 3 − 188 231 x 1 x 4 + 292 ( x 2 + − 2 − 1 ) 1155 x 3 x 4 � � − 467 693 + 842 693 x 1 x 2 + 908 693 x 1 x 4 + 292 ( x 2 + 693 x 2 x 4 3 − 1 ) � � − 68 693 − 376 693 x 1 x 2 + 1816 3465 x 1 x 3 + 584 ( x 2 + 4 − 1 ) 3465 x 2 x 3 � 155 693 x 2 + 467 1 3465 x 3 + 34 693 x 4 − 842 + 693 x 1 + 3465 x 1 x 2 x 3 � + 188 693 x 1 x 2 x 4 − 908 3465 x 1 x 3 x 4 − 292 ( x 1 + 3x 2 + 5x 3 + 2x 4 ) . 3465 x 2 x 3 x 4 − 51975 = (1 + 3 + 5 + 2)( − 1 + 3 + 5 + 2)(1 − 3 + 5 + 2)(1 + 3 − 5 + 2) ( − 1 − 3 + 5 + 2)( − 1 + 3 − 5 + 2)(1 − 3 − 5 + 2)( − 1 − 3 − 5 + 2) . Via Cramer’s rule, we see that the unknown b 4 is equal to b 4 = − 2550 − 51975 = 34 693 . Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  65. Definition of Graph Coloring Graph coloring: Given a graph G , and an integer k , can the vertices be colored with k colors in such a way that no two adjacent vertices are the same color? Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  66. Definition of Graph Coloring Graph coloring: Given a graph G , and an integer k , can the vertices be colored with k colors in such a way that no two adjacent vertices are the same color? Petersen Graph: 3-colorable Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  67. Graph 3-Coloring as a System of Polynomial Equations over C (D. Bayer) one variable per vertex : x 1 , . . . , x n Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  68. Graph 3-Coloring as a System of Polynomial Equations over C (D. Bayer) one variable per vertex : x 1 , . . . , x n vertex polynomials: For every vertex i = 1 , . . . , n , x 3 i − 1 = 0 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  69. Graph 3-Coloring as a System of Polynomial Equations over C (D. Bayer) one variable per vertex : x 1 , . . . , x n vertex polynomials: For every vertex i = 1 , . . . , n , x 3 i − 1 = 0 edge polynomials: For every edge ( i , j ) ∈ E ( G ), x 2 i + x i x j + x 2 j = 0 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  70. Graph 3-Coloring as a System of Polynomial Equations over C (D. Bayer) one variable per vertex : x 1 , . . . , x n vertex polynomials: For every vertex i = 1 , . . . , n , x 3 i − 1 = 0 edge polynomials: For every edge ( i , j ) ∈ E ( G ), x 3 i − x 3 j = x 2 i + x i x j + x 2 j = 0 x i − x j Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  71. Graph 3-Coloring as a System of Polynomial Equations over C (D. Bayer) one variable per vertex : x 1 , . . . , x n vertex polynomials: For every vertex i = 1 , . . . , n , x 3 i − 1 = 0 edge polynomials: For every edge ( i , j ) ∈ E ( G ), x 3 i − x 3 j = x 2 i + x i x j + x 2 j = 0 x i − x j Theorem: Let G be a graph encoded as the above ( n + m ) system of equations. Then this system has a solution if and only if G is 3-colorable. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  72. Petersen Graph = ⇒ System of Polynomial Equations Figure: Is the Petersen graph 3-colorable? x 3 0 − 1 = 0 , x 3 x 2 0 + x 0 x 1 + x 2 1 = 0 , x 2 0 + x 0 x 4 + x 2 1 − 1 = 0 , 4 = 0 x 3 2 − 1 = 0 , x 3 x 2 0 + x 0 x 5 + x 2 5 = 0 , x 2 1 + x 1 x 2 + x 2 3 − 1 = 0 , 2 = 0 x 3 4 − 1 = 0 , x 3 x 2 1 + x 1 x 6 + x 2 6 = 0 , x 2 2 + x 2 x 3 + x 2 5 − 1 = 0 , 3 = 0 x 3 6 − 1 = 0 , x 3 7 − 1 = 0 , · · · · · · · · · · · · x 3 8 − 1 = 0 , x 3 x 2 6 + x 6 x 8 + x 2 8 = 0 , x 2 7 + x 7 x 9 + x 2 9 − 1 = 0 , 9 = 0 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  73. Where is the Infinite Family of Graphs that Grow over C ? 4 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  74. Where is the Infinite Family of Graphs that Grow over C ? 4 Flower, Kneser, Gr¨ otzsch, Jin, Mycielski graphs have degree 4. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  75. Where is the Infinite Family of Graphs that Grow over C ? 4 Flower, Kneser, Gr¨ otzsch, Jin, Mycielski graphs have degree 4. Theorem: Every Nullstellensatz certificate of a non-3-colorable graph has degree at least four. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  76. Where is the Infinite Family of Graphs that Grow over C ? 4 Flower, Kneser, Gr¨ otzsch, Jin, Mycielski graphs have degree 4. Theorem: Every Nullstellensatz certificate of a non-3-colorable graph has degree at least four. Theorem: For n ≥ 4, a minimum-degree Nullstellensatz certificate of non-3-colorability for cliques and odd wheels has degree exactly four. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  77. Graph 3-Coloring as a System of Polynomial Equations over F 2 (inspired by Bayer) one variable per vertex : x 1 , . . . , x n vertex polynomials: For every vertex i = 1 , . . . , n , x 3 i + 1 = 0 edge polynomials: For every edge ( i , j ) ∈ E ( G ), x 2 i + x i x j + x 2 j = 0 Theorem: Let G be a graph encoded as the above ( n + m ) system of equations. Then this system has a solution if and only if G is 3-colorable. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  78. Where is the Infinite Family of Graphs that Grow over F 2 ? 1 Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  79. Where is the Infinite Family of Graphs that Grow over F 2 ? 1 Theorem: Every Nullstellensatz certificate of a non-3-colorable graph has degree at least one. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  80. Where is the Infinite Family of Graphs that Grow over F 2 ? 1 Theorem: Every Nullstellensatz certificate of a non-3-colorable graph has degree at least one. Theorem: For n ≥ 4, a minimum-degree Nullstellensatz certificate of non-3-colorability for cliques and odd wheels has degree exactly one. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  81. Experimental results for NulLA 3-colorability Graph vertices edges rows cols deg sec Mycielski 7 95 755 64,281 71,726 Mycielski 9 383 7,271 2,477,931 2,784,794 Mycielski 10 767 22,196 15,270,943 17,024,333 (8 , 3)-Kneser 56 280 15,737 15,681 (10 , 4)-Kneser 210 1,575 349,651 330,751 (12 , 5)-Kneser 792 8,316 7,030,585 6,586,273 (13 , 5)-Kneser 1,287 36,036 45,980,650 46,378,333 1-Insertions 5 202 1,227 268,049 247,855 2-Insertions 5 597 3,936 2,628,805 2,349,793 3-Insertions 5 1,406 9,695 15,392,209 13,631,171 ash331GPIA 662 4,185 3,147,007 2,770,471 ash608GPIA 1,216 7,844 10,904,642 9,538,305 ash958GPIA 1,916 12,506 27,450,965 23,961,497 Table: Graphs without 4-cliques. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

  82. Experimental results for NulLA 3-colorability Graph vertices edges rows cols deg sec Mycielski 7 95 755 64,281 71,726 1 Mycielski 9 383 7,271 2,477,931 2,784,794 1 Mycielski 10 767 22,196 15,270,943 17,024,333 1 (8 , 3)-Kneser 56 280 15,737 15,681 1 (10 , 4)-Kneser 210 1,575 349,651 330,751 1 (12 , 5)-Kneser 792 8,316 7,030,585 6,586,273 1 (13 , 5)-Kneser 1,287 36,036 45,980,650 46,378,333 1 1-Insertions 5 202 1,227 268,049 247,855 1 2-Insertions 5 597 3,936 2,628,805 2,349,793 1 3-Insertions 5 1,406 9,695 15,392,209 13,631,171 1 ash331GPIA 662 4,185 3,147,007 2,770,471 1 ash608GPIA 1,216 7,844 10,904,642 9,538,305 1 ash958GPIA 1,916 12,506 27,450,965 23,961,497 1 Table: Graphs without 4-cliques. Susan Margulies, US Naval Academy NulLA and Combinatorial Infeasibility

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Linear Algebra Linear algebra has become as basic and as applicable as calculus, and

Linear Algebra Linear algebra has become as basic and as applicable as calculus, and

Mathematical Tools for Neural and Cognitive Science Fall semester, 2018 Section 1: Linear Algebra Linear Algebra Linear algebra has become as basic and as applicable as calculus, and fortunately it is easier - Gilbert Strang, Linear

359 views • 10 slides
Hilbert Functions in Algebra and Geometry Alexandra Seceleanu Department of Mathematics

Hilbert Functions in Algebra and Geometry Alexandra Seceleanu Department of Mathematics

Hilbert Functions in Algebra and Geometry Alexandra Seceleanu Department of Mathematics University of NebraskaLincoln GWCAWMMG workshop April 13, 2019 Outline What is a Hilbert function? Hilberts Theorem Classification of Hilbert

1.07k views • 41 slides
Linear Algebra Review Leila Wehbe January 29, 2013 Leila Wehbe Linear Algebra Review Metrics

Linear Algebra Review Leila Wehbe January 29, 2013 Leila Wehbe Linear Algebra Review Metrics

Metrics Vector Spaces Banach Spaces Hilbert Space Matrices Linear Algebra Review Leila Wehbe January 29, 2013 Leila Wehbe Linear Algebra Review Metrics Vector Spaces Banach Spaces Hilbert Space Matrices Table of contents Metrics

439 views • 25 slides
Chapter 1 What is Linear Algebra? Chapter 1 What is Linear Algebra? The study of linear

Chapter 1 What is Linear Algebra? Chapter 1 What is Linear Algebra? The study of linear

Chapter 1 What is Linear Algebra? Chapter 1 What is Linear Algebra? The study of linear functions. The word linear means straight or flat . y = 0 + 1 x Linear functions involve only addition and scalar multiplication. Chapter 1 Higher

1.2k views • 56 slides
Linear algebra explained in four pages Excerpt from the N O BULLSHIT GUIDE TO LINEAR ALGEBRA by

Linear algebra explained in four pages Excerpt from the N O BULLSHIT GUIDE TO LINEAR ALGEBRA by

Linear algebra explained in four pages Excerpt from the N O BULLSHIT GUIDE TO LINEAR ALGEBRA by Ivan Savov Abstract This document will review the fundamental ideas of linear algebra. B. Matrix operations We will learn about matrices, matrix

333 views • 4 slides
Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations

Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations

Peter Latham, October 7, 2014 1 Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations of the form A x = y , (1) { lin_eq } which is written in terms of components as A ij x j = y j . (2) j

477 views • 11 slides
Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations

Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations

Peter Latham, September 30, 2014 1 Math you need to know 1 Linear algebra Linear algebra is mainly concerned with solving equations of the form A x = y , (1) { lin_eq } which is written in terms of components as A ij x j = y j . (2)

273 views • 6 slides
Matrices Basic Linear Algebra Overview Lecture will cover why matrices and linear algebra

Matrices Basic Linear Algebra Overview Lecture will cover why matrices and linear algebra

Matrices Basic Linear Algebra Overview Lecture will cover why matrices and linear algebra are so important basic terminology Gauss-Jordan elimination LU factorisation error estimation libraries Basic Linear Algebra 2

564 views • 27 slides
MATRICES AND LINEAR ALGEBRA Linear Algebra Matrix manipulation is the original essence of

MATRICES AND LINEAR ALGEBRA Linear Algebra Matrix manipulation is the original essence of

MATRICES AND LINEAR ALGEBRA Linear Algebra Matrix manipulation is the original essence of Matlab; hence the name MATrix LABoratory. In this section we will cover the basics of linear algebra, the ways of using Matlab in this context, and

1.08k views • 28 slides
Chapter 4: Vectors, Matrices, and Linear Algebra Scott Owen & Greg Corrado Linear Algebra is

Chapter 4: Vectors, Matrices, and Linear Algebra Scott Owen & Greg Corrado Linear Algebra is

Chapter 4: Vectors, Matrices, and Linear Algebra Scott Owen & Greg Corrado Linear Algebra is strikingly similar to the algebra you learned in high school, except that in the place of ordinary single numbers, it deals with vectors. Many of

650 views • 11 slides
Lecture 14: Dense Linear Algebra David Bindel 18 Oct 2010 Where we are This week: dense

Lecture 14: Dense Linear Algebra David Bindel 18 Oct 2010 Where we are This week: dense

Lecture 14: Dense Linear Algebra David Bindel 18 Oct 2010 Where we are This week: dense linear algebra Next week: sparse linear algebra Numerical linear algebra in a nutshell Basic problems Linear systems: Ax = b Least

539 views • 36 slides
Linear Algebra 1: Computing canonical forms in exact linear algebra Clment P ERNET ,

Linear Algebra 1: Computing canonical forms in exact linear algebra Clment P ERNET ,

Linear Algebra 1: Computing canonical forms in exact linear algebra Clment P ERNET , LIG/INRIA-MOAIS, Grenoble Universit, France ECRYPT II: Summer School on Tools, Mykonos, Grece, June 1st, 2012 Introduction : matrix normal forms Given a

1.64k views • 136 slides
Graphical Linear Algebra a specification language for linear algebra Pawel Sobocinski based on

Graphical Linear Algebra a specification language for linear algebra Pawel Sobocinski based on

Graphical Linear Algebra a specification language for linear algebra Pawel Sobocinski based on joint work with Filippo Bonchi and Fabio Zanasi CONCUR My first CS conference: Concur 2001, Aalborg My first talk: EXPRESS 2002, Brno (a

839 views • 56 slides
20/11/2019 20/11/2019 2 Outline Introduction to Linear Algebra Algebra Solving

20/11/2019 20/11/2019 2 Outline Introduction to Linear Algebra Algebra Solving

20/11/2019 20/11/2019 2 Outline Introduction to Linear Algebra Algebra Solving equations Matrices Leonardo de Knegt Types of matrices Matrix operations Systems of linear equations Linear regression State-space

333 views • 6 slides
Essence of Linear Algebra Linear combinations and Span Some cool intuitions The box game:

Essence of Linear Algebra Linear combinations and Span Some cool intuitions The box game:

Essence of Linear Algebra Shreedhar Kodate The Geometry of Linear Equations Vectors and Basis vectors Essence of Linear Algebra Linear combinations and Span Some cool intuitions The box game: Matrices Elimination and Multiplication,

424 views • 41 slides
Linear algebra and analysis recalls Lectures for PHD course on Numerical optimization Enrico

Linear algebra and analysis recalls Lectures for PHD course on Numerical optimization Enrico

Linear algebra and analysis recalls Lectures for PHD course on Numerical optimization Enrico Bertolazzi DIMS Universit a di Trento November 21 December 14, 2011 Linear algebra and analysis recalls 1 / 30 Outline Linear algebra 1

348 views • 30 slides
Early Learners on the autism spectrum Marleen Westerveld, PhD Project Team Griffith University:

Early Learners on the autism spectrum Marleen Westerveld, PhD Project Team Griffith University:

Literacy Profiles and Literacy Predictors for Early Learners on the autism spectrum Marleen Westerveld, PhD Project Team Griffith University: Sydney Childrens Hospital at Westmead: Dr David Trembath Dr Antoinette Hodge Dr

693 views • 35 slides
BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks ICPR 2016 Surat

BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks ICPR 2016 Surat

BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks ICPR 2016 Surat Teerapittayanon Brad McDanel H. T. Kung Harvard John A. Paulson School of Engineering and Applied Sciences 1 outline Motivation and Background Trend

1.49k views • 23 slides
... - Chkago . Pittsburgh Now York Gty and c .... HoriMmCoIot.do ento ~ Refueling

... - Chkago . Pittsburgh Now York Gty and c .... HoriMmCoIot.do ento ~ Refueling

.~-"~ ~. t W~ ~WoblMt. . 1oW._h.I.~m E.utm~. ~"8q f'onciKo ~ . ~ton s.n Y~_T_ ~in C_oIltod~ w~ . . 'cU~ ~ ~ _ ~ v~ Clean Cities: A voluntary, locally-based governmenUindustry partnership 87 Active Coalitions in 46

295 views • 13 slides
American International Group, Inc. Third Quarter 2012 Results Conference Call Presentation

American International Group, Inc. Third Quarter 2012 Results Conference Call Presentation

American International Group, Inc. Third Quarter 2012 Results Conference Call Presentation November 2 nd , 2012 Cautionary Statement Regarding Projections and Other Information About Future Events This document and the remarks made within this

551 views • 19 slides
The Welfare Impact of Global Migration in the OECD Countries A. Aubry and M. Burzy ski

The Welfare Impact of Global Migration in the OECD Countries A. Aubry and M. Burzy ski

The Welfare Impact of Global Migration in the OECD Countries A. Aubry and M. Burzy ski Discussion Paper 2013-35 r t rt t

646 views • 41 slides
IPA Assembly 2017 Presentation for Chapter Creator God, you are luring us into a future more

IPA Assembly 2017 Presentation for Chapter Creator God, you are luring us into a future more

IPA Assembly 2017 Presentation for Chapter Creator God, you are luring us into a future more expansive than our minds can conceive. Grace us and make space in the womb of our hearts for the possibilities you offer us. Unsettle us! Disturb

626 views • 51 slides
THE MOTHER OF THE QURAAN Session B Breakdown In Introduction

THE MOTHER OF THE QURAAN Session B Breakdown In Introduction

Mufti Abdul Wahab Waheed THE MOTHER OF THE QURAAN Session B Breakdown In Introduction Effect o of t f the Q Quran Virtues o of S f Surah F Fatiha Names o of t f the S Surah Is

653 views • 49 slides
2019 Probate Academy Dr. Gerry W. Beyer Governor Preston E. Smith Regents Professor of Law

2019 Probate Academy Dr. Gerry W. Beyer Governor Preston E. Smith Regents Professor of Law

2019 Probate Academy Dr. Gerry W. Beyer Governor Preston E. Smith Regents Professor of Law Texas Tech University School of Law 1 2 Chief Justice United States Supreme Court Warren Burger 3 President of the United States Abraham Lincoln 4

923 views • 65 slides

More recommend