Today's Web-enabled deluge of electronic data calls for automated methods of data analysis. Machine learning provides these, developing methods that can automatically detect patterns in data and then use the uncovered patterns to predict future data. This textbook offers a comprehensive and self-contained introduction to the field of machine learning, based on a unified, probabilistic approach. The coverage combines breadth and depth, offering necessary background material on such topics as probability, optimization, and linear algebra as well as discussion of recent developments in the field, including conditional random fields, L1 regularization, and deep learning. The book is written in an informal, accessible style, complete with pseudo-code for the most important algorithms. All topics are copiously illustrated with color images and worked examples drawn from such application domains as biology, text processing, computer vision, and robotics. Rather than providing a cookbook of different heuristic methods, the book stresses a principled model-based approach, often using the language of graphical models to specify models in a concise and intuitive way. Almost all the models described have been implemented in a MATLAB software package--PMTK (probabilistic modeling toolkit)--that is freely available online. The book is suitable for upper-level undergraduates with an introductory-level college math background and beginning graduate students.

Machine Learning : A Probabilistic Perspective

We describe a Bayesian approach for learning Bayesian networks from a combination of prior knowledge and statistical data. First and foremost, we develop a methodology for assessing informative priors needed for learning. Our approach is derived from a set of assumptions made previously as well as the assumption of likelihood equivalence, which says that data should not help to discriminate network structures that represent the same assertions of conditional independence. We show that likelihood equivalence when combined with previously made assumptions implies that the user's priors for network parameters can be encoded in a single Bayesian network for the next case to be seen—a prior network—and a single measure of confidence for that network. Second, using these priors, we show how to compute the relative posterior probabilities of network structures given data. Third, we describe search methods for identifying network structures with high posterior probabilities. We describe polynomial algorithms for finding the highest-scoring network structures in the special case where every node has at most k e 1 parent. For the general case (k > 1), which is NP-hard, we review heuristic search algorithms including local search, iterative local search, and simulated annealing. Finally, we describe a methodology for evaluating Bayesian-network learning algorithms, and apply this approach to a comparison of various approaches.

/pdf/learning-bayesian-networks-the-combination-of-knowledge-and-5401ouuevj.pdf

Learning Bayesian Networks: The Combination of Knowledge and Statistical Data

is one of the most recognizablecharacters in 20th century cinema. HAL is an artiﬁcial agent capable of such advancedlanguage behavior as speaking and understanding English, and at a crucial moment inthe plot, even reading lips. It is now clear that HAL’s creator, Arthur C. Clarke, wasa little optimistic in predicting when an artiﬁcial agent such as HAL would be avail-able. But just how far off was he? What would it take to create at least the language-relatedpartsofHAL?WecallprogramslikeHALthatconversewithhumansinnatural

/pdf/speech-and-language-processing-1xjlmt8snu.pdf

Speech and Language Processing

problems To understand the class of polynomial-time solvable problems, we must first have a formal notion of what a "problem" is. We define an abstract problem Q to be a binary relation on a set I of problem instances and a set S of problem solutions. For example, an instance for SHORTEST-PATH is a triple consisting of a graph and two vertices. A solution is a sequence of vertices in the graph, with perhaps the empty sequence denoting that no path exists. The problem SHORTEST-PATH itself is the relation that associates each instance of a graph and two vertices with a shortest path in the graph that connects the two vertices. Since shortest paths are not necessarily unique, a given problem instance may have more than one solution. This formulation of an abstract problem is more general than is required for our purposes. As we saw above, the theory of NP-completeness restricts attention to decision problems: those having a yes/no solution. In this case, we can view an abstract decision problem as a function that maps the instance set I to the solution set {0, 1}. For example, a decision problem related to SHORTEST-PATH is the problem PATH that we saw earlier. If i = G, u, v, k is an instance of the decision problem PATH, then PATH(i) = 1 (yes) if a shortest path from u to v has at most k edges, and PATH(i) = 0 (no) otherwise. Many abstract problems are not decision problems, but rather optimization problems, in which some value must be minimized or maximized. As we saw above, however, it is usually a simple matter to recast an optimization problem as a decision problem that is no harder. Encodings If a computer program is to solve an abstract problem, problem instances must be represented in a way that the program understands. An encoding of a set S of abstract objects is a mapping e from S to the set of binary strings. For example, we are all familiar with encoding the natural numbers N = {0, 1, 2, 3, 4,...} as the strings {0, 1, 10, 11, 100,...}. Using this encoding, e(17) = 10001. Anyone who has looked at computer representations of keyboard characters is familiar with either the ASCII or EBCDIC codes. In the ASCII code, the encoding of A is 1000001. Even a compound object can be encoded as a binary string by combining the representations of its constituent parts. Polygons, graphs, functions, ordered pairs, programs-all can be encoded as binary strings. Thus, a computer algorithm that "solves" some abstract decision problem actually takes an encoding of a problem instance as input. We call a problem whose instance set is the set of binary strings a concrete problem. We say that an algorithm solves a concrete problem in time O(T (n)) if, when it is provided a problem instance i of length n = |i|, the algorithm can produce the solution in O(T (n)) time. A concrete problem is polynomial-time solvable, therefore, if there exists an algorithm to solve it in time O(n) for some constant k. We can now formally define the complexity class P as the set of concrete decision problems that are polynomial-time solvable. We can use encodings to map abstract problems to concrete problems. Given an abstract decision problem Q mapping an instance set I to {0, 1}, an encoding e : I → {0, 1}* can be used to induce a related concrete decision problem, which we denote by e(Q). If the solution to an abstract-problem instance i I is Q(i) {0, 1}, then the solution to the concreteproblem instance e(i) {0, 1}* is also Q(i). As a technicality, there may be some binary strings that represent no meaningful abstract-problem instance. For convenience, we shall assume that any such string is mapped arbitrarily to 0. Thus, the concrete problem produces the same solutions as the abstract problem on binary-string instances that represent the encodings of abstract-problem instances. We would like to extend the definition of polynomial-time solvability from concrete problems to abstract problems by using encodings as the bridge, but we would like the definition to be independent of any particular encoding. That is, the efficiency of solving a problem should not depend on how the problem is encoded. Unfortunately, it depends quite heavily on the encoding. For example, suppose that an integer k is to be provided as the sole input to an algorithm, and suppose that the running time of the algorithm is Θ(k). If the integer k is provided in unary-a string of k 1's-then the running time of the algorithm is O(n) on length-n inputs, which is polynomial time. If we use the more natural binary representation of the integer k, however, then the input length is n = ⌊lg k⌋ + 1. In this case, the running time of the algorithm is Θ (k) = Θ(2), which is exponential in the size of the input. Thus, depending on the encoding, the algorithm runs in either polynomial or superpolynomial time. The encoding of an abstract problem is therefore quite important to our under-standing of polynomial time. We cannot really talk about solving an abstract problem without first specifying an encoding. Nevertheless, in practice, if we rule out "expensive" encodings such as unary ones, the actual encoding of a problem makes little difference to whether the problem can be solved in polynomial time. For example, representing integers in base 3 instead of binary has no effect on whether a problem is solvable in polynomial time, since an integer represented in base 3 can be converted to an integer represented in base 2 in polynomial time. We say that a function f : {0, 1}* → {0,1}* is polynomial-time computable if there exists a polynomial-time algorithm A that, given any input x {0, 1}*, produces as output f (x). For some set I of problem instances, we say that two encodings e1 and e2 are polynomially related if there exist two polynomial-time computable functions f12 and f21 such that for any i I , we have f12(e1(i)) = e2(i) and f21(e2(i)) = e1(i). That is, the encoding e2(i) can be computed from the encoding e1(i) by a polynomial-time algorithm, and vice versa. If two encodings e1 and e2 of an abstract problem are polynomially related, whether the problem is polynomial-time solvable or not is independent of which encoding we use, as the following lemma shows. Lemma 34.1 Let Q be an abstract decision problem on an instance set I , and let e1 and e2 be polynomially related encodings on I . Then, e1(Q) P if and only if e2(Q) P. Proof We need only prove the forward direction, since the backward direction is symmetric. Suppose, therefore, that e1(Q) can be solved in time O(nk) for some constant k. Further, suppose that for any problem instance i, the encoding e1(i) can be computed from the encoding e2(i) in time O(n) for some constant c, where n = |e2(i)|. To solve problem e2(Q), on input e2(i), we first compute e1(i) and then run the algorithm for e1(Q) on e1(i). How long does this take? The conversion of encodings takes time O(n), and therefore |e1(i)| = O(n), since the output of a serial computer cannot be longer than its running time. Solving the problem on e1(i) takes time O(|e1(i)|) = O(n), which is polynomial since both c and k are constants. Thus, whether an abstract problem has its instances encoded in binary or base 3 does not affect its "complexity," that is, whether it is polynomial-time solvable or not, but if instances are encoded in unary, its complexity may change. In order to be able to converse in an encoding-independent fashion, we shall generally assume that problem instances are encoded in any reasonable, concise fashion, unless we specifically say otherwise. To be precise, we shall assume that the encoding of an integer is polynomially related to its binary representation, and that the encoding of a finite set is polynomially related to its encoding as a list of its elements, enclosed in braces and separated by commas. (ASCII is one such encoding scheme.) With such a "standard" encoding in hand, we can derive reasonable encodings of other mathematical objects, such as tuples, graphs, and formulas. To denote the standard encoding of an object, we shall enclose the object in angle braces. Thus, G denotes the standard encoding of a graph G. As long as we implicitly use an encoding that is polynomially related to this standard encoding, we can talk directly about abstract problems without reference to any particular encoding, knowing that the choice of encoding has no effect on whether the abstract problem is polynomial-time solvable. Henceforth, we shall generally assume that all problem instances are binary strings encoded using the standard encoding, unless we explicitly specify the contrary. We shall also typically neglect the distinction between abstract and concrete problems. The reader should watch out for problems that arise in practice, however, in which a standard encoding is not obvious and the encoding does make a difference. A formal-language framework One of the convenient aspects of focusing on decision problems is that they make it easy to use the machinery of formal-language theory. It is worthwhile at this point to review some definitions from that theory. An alphabet Σ is a finite set of symbols. A language L over Σ is any set of strings made up of symbols from Σ. For example, if Σ = {0, 1}, the set L = {10, 11, 101, 111, 1011, 1101, 10001,...} is the language of binary representations of prime numbers. We denote the empty string by ε, and the empty language by Ø. The language of all strings over Σ is denoted Σ*. For example, if Σ = {0, 1}, then Σ* = {ε, 0, 1, 00, 01, 10, 11, 000,...} is the set of all binary strings. Every language L over Σ is a subset of Σ*. There are a variety of operations on languages. Set-theoretic operations, such as union and intersection, follow directly from the set-theoretic definitions. We define the complement of L by . The concatenation of two languages L1 and L2 is the language L = {x1x2 : x1 L1 and x2 L2}. The closure or Kleene star of a language L is the language L*= {ε} L L L ···, where Lk is the language obtained by

Introduction to Algorithms, Second Edition

In this paper we develop a new data structure for implementing heaps (priority queues). Our structure, Fibonacci heaps (abbreviated F-heaps), extends the binomial queues proposed by Vuillemin and studied further by Brown. F-heaps support arbitrary deletion from an n-item heap in O(log n) amortized time and all other standard heap operations in O(1) amortized time. Using F-heaps we are able to obtain improved running times for several network optimization algorithms. In particular, we obtain the following worst-case bounds, where n is the number of vertices and m the number of edges in the problem graph: O(n log n + m) for the single-source shortest path problem with nonnegative edge lengths, improved from O(mlog(m/n+2)n);O(n2log n + nm) for the all-pairs shortest path problem, improved from O(nm log(m/n+2)n);O(n2log n + nm) for the assignment problem (weighted bipartite matching), improved from O(nmlog(m/n+2)n);O(mβ(m, n)) for the minimum spanning tree problem, improved from O(mlog log(m/n+2)n); where β(m, n) = min {i | log(i)n ≤ m/n}. Note that β(m, n) ≤ log*n if m ≥ n.Of these results, the improved bound for minimum spanning trees is the most striking, although all the results give asymptotic improvements for graphs of appropriate densities.

/pdf/fibonacci-heaps-and-their-uses-in-improved-network-3s6u3e39gv.pdf

Fibonacci heaps and their uses in improved network optimization algorithms

Recently, Fredman and Tarjan invented a new, especially efficient form of heap (priority queue). Their data structure, theFibonacci heap (or F-heap) supports arbitrary deletion inO(logn) amortized time and other heap operations inO(1) amortized time. In this paper we use F-heaps to obtain fast algorithms for finding minimum spanning trees in undirected and directed graphs. For an undirected graph containingn vertices andm edges, our minimum spanning tree algorithm runs inO(m logβ (m, n)) time, improved fromO(mβ(m, n)) time, whereβ(m, n)=min {i|log(i)
 n ≦m/n}. Our minimum spanning tree algorithm for directed graphs runs inO(n logn + m) time, improved fromO(n log n +m log log log(m/n+2)
 n). Both algorithms can be extended to allow a degree constraint at one vertex.

Efficient algorithms for finding minimum spanning trees in undirected and directed graphs

The (component) merging problem is a new graph problem. Versions of this problem appear as bottlenecks in various graph algorithms. A new data structure solves this problem efficiently, and two special cases of the problem have even more efficient solutions based on other data structures. The performance of the data structures is sped up by introducing a new algorithmic tool called packets.The algorithms that use these solutions to the component merging problem also exploit new properties of two existing data structures. Specifically, B-trees can be used simultaneously as a priority queue and a concatenable queue. Similarly, F-heaps support some kinds of split operations with no loss of efficiency.An immediate application of the solution to the simplest version of the merging problem is an O(t(m, n)) algorithm for finding minimum spanning trees in undirected graphs without using F-heaps, where t(m, n) = mlog2log2logdn, the graph has n vertices and m edges, and d = max(m/n, 2). Packets also improve the F-heap minimum spanning tree algorithm, giving the fastest algorithm currently known for this problem.The efficient solutions to the merging problem and the new observation about F-heaps lead to an O(n(t(m, n) + nlogn)) algorithm for finding a maximum weighted matching in general graphs. This settles an open problem posed by Tarjan [ 15, p. 123], where the weaker bound of O(nm log (n2/m)) was conjectured.

Efficient implementation of graph algorithms using contraction

A new parallel algorithm for the maximal independent set problem (MIS) is constructed. It runs in O(log4 n) time when implemented on a linear number of EREW-processors. This is the first deterministic algorithm for MIS whose running time is polylogarithmic and whose processor-time product is optimal up to a polylogarithmic factor.

A new parallel algorithm for the maximal independent set problem

A new parallel algorithm for the maximal independent set problem is constructed. It runs in $O(\log ^4 n)$ time when implemented on a linear number of EREW-processors. This is the first deterministic algorithm for the maximal independent set problem (MIS) whose running time is polylogarithmic and whose processor-time product is optimal up to a polylogarithmic factor.

The problem of constructing in parallel a maximal independent set of a given graph is considered. A new deterministic NC-algorithm implemented in the EREW PRAM model is presented. On graphs with n vertices and m edges, it uses $O((n + m)/\log n)$ processors and runs in $O(\log^3 n)$ time. This reduces by a factor of $\log n$ both the running time and the processor count of the previously fastest deterministic algorithm that solves the problem using a linear number of processors.

/pdf/constructing-a-maximal-independent-set-in-parallel-4xuwffdt6i.pdf

Thomas H. Spencer

Papers

Efficient algorithms for finding minimum spanning trees in undirected and directed graphs

Efficient implementation of graph algorithms using contraction

A new parallel algorithm for the maximal independent set problem

A new parallel algorithm for the maximal independent set problem

Constructing a maximal independent set in parallel