Advances in training models with log-linear structures, with topics including variable selection, the geometry of neural nets, and applications.

Log-linear models play a key role in modern big data and machine learning applications. From simple binary classification models through partition functions, conditional random fields, and neural nets, log-linear structure is closely related to performance in certain applications and influences fitting techniques used to train models. This volume covers recent advances in training models with log-linear structures, covering the underlying geometry, optimization techniques, and multiple applications. The first chapter shows readers the inner workings of machine learning, providing insights into the geometry of log-linear and neural net models. The other chapters range from introductory material to optimization techniques to involved use cases. The book, which grew out of a NIPS workshop, is suitable for graduate students doing research in machine learning, in particular deep learning, variable selection, and applications to speech recognition. The contributors come from academia and industry, allowing readers to view the field from both perspectives.

ContributorsAleksandr Aravkin, Avishy Carmi, Guillermo A. Cecchi, Anna Choromanska, Li Deng, Xinwei Deng, Jean Honorio, Tony Jebara, Huijing Jiang, Dimitri Kanevsky, Brian Kingsbury, Fabrice Lambert, Aurélie C. Lozano, Daniel Moskovich, Yuriy S. Polyakov, Bhuvana Ramabhadran, Irina Rish, Dimitris Samaras, Tara N. Sainath, Hagen Soltau, Serge F. Timashev, Ewout van den Berg

Machine Learning: Discriminative and Generative covers the main contemporary themes and tools in machine learning ranging from Bayesian probabilistic models to discriminative support-vector machines. However, unlike previous books that only discuss these rather different approaches in isolation, it bridges the two schools of thought together within a common framework, elegantly connecting their various theories and making one common big-picture. Also, this bridge brings forth new hybrid discriminative-generative tools that combine the strengths of both camps. This book serves multiple purposes as well. The framework acts as a scientific breakthrough, fusing the areas of generative and discriminative learning and will be of interest to many researchers. However, as a conceptual breakthrough, this common framework unifies many previously unrelated tools and techniques and makes them understandable to a larger portion of the public. This gives the more practical-minded engineer, student and the industrial public an easy-access and more sensible road map into the world of machine learning.

Machine Learning: Discriminative and Generative is designed for an audience composed of researchers & practitioners in industry and academia. The book is also suitable as a secondary text for graduate-level students in computer science and engineering.

Advances in training models with log-linear structures, with topics including variable selection, the geometry of neural nets, and applications.

Log-linear models play a key role in modern big data and machine learning applications. From simple binary classification models through partition functions, conditional random fields, and neural nets, log-linear structure is closely related to performance in certain applications and influences fitting techniques used to train models. This volume covers recent advances in training models with log-linear structures, covering the underlying geometry, optimization techniques, and multiple applications. The first chapter shows readers the inner workings of machine learning, providing insights into the geometry of log-linear and neural net models. The other chapters range from introductory material to optimization techniques to involved use cases. The book, which grew out of a NIPS workshop, is suitable for graduate students doing research in machine learning, in particular deep learning, variable selection, and applications to speech recognition. The contributors come from academia and industry, allowing readers to view the field from both perspectives.

ContributorsAleksandr Aravkin, Avishy Carmi, Guillermo A. Cecchi, Anna Choromanska, Li Deng, Xinwei Deng, Jean Honorio, Tony Jebara, Huijing Jiang, Dimitri Kanevsky, Brian Kingsbury, Fabrice Lambert, Aurélie C. Lozano, Daniel Moskovich, Yuriy S. Polyakov, Bhuvana Ramabhadran, Irina Rish, Dimitris Samaras, Tara N. Sainath, Hagen Soltau, Serge F. Timashev, Ewout van den Berg

Machine Learning: Discriminative and Generative covers the main contemporary themes and tools in machine learning ranging from Bayesian probabilistic models to discriminative support-vector machines. However, unlike previous books that only discuss these rather different approaches in isolation, it bridges the two schools of thought together within a common framework, elegantly connecting their various theories and making one common big-picture. Also, this bridge brings forth new hybrid discriminative-generative tools that combine the strengths of both camps. This book serves multiple purposes as well. The framework acts as a scientific breakthrough, fusing the areas of generative and discriminative learning and will be of interest to many researchers. However, as a conceptual breakthrough, this common framework unifies many previously unrelated tools and techniques and makes them understandable to a larger portion of the public. This gives the more practical-minded engineer, student and the industrial public an easy-access and more sensible road map into the world of machine learning.

Machine Learning: Discriminative and Generative is designed for an audience composed of researchers & practitioners in industry and academia. The book is also suitable as a secondary text for graduate-level students in computer science and engineering.