Krishnan Raghavachari<\/h3>\nDistinguished Professor Indiana University<\/h4>\nAchieving Coupled Cluster Accuracy at DFT Cost: Novel Ideas from Molecular Fragmentation, Error Cancellation and Machine Learning<\/h5>\n
Originally from India, Prof. Krishnan Raghavachari received his B.Sc. in Chemistry from Madras University (1973) and his M.Sc. in Chemistry from Indian Institute of Technology, Madras (1975). He received his Ph.D. from Carnegie Mellon University in 1981 working with legendary Professor John Pople (Nobel Laureate in Chemistry, 1998). He joined Bell Laboratories as a research scientist in 1981 and later received the Distinguished Researcher Award in 1987. He joined Indiana University as a Professor of Chemistry in 2002 and was appointed as a Distinguished Professor in 2014.<\/p>\n
Professor Raghavachari is well known for his work on the development of electron correlation theories and is the principal author of the highly successful CCSD(T) technique (1989), often called \u201cthe Gold Standard of Quantum Chemistry\u201d. Research in his group focuses on the development of accurate and insightful computational models in quantum chemistry and for pioneering their use in forefront applications ranging from semiconductor surface chemistry to computer-aided drug design. He is a pioneer in the development of fragmentation-based methods in quantum chemistry that enable quantitatively accurate studies on large molecules. He has expanded his research profile in recent years to include machine learning in quantum chemistry, and computer-aided design of fluorescent materials and receptors. He has co-authored more than 450 publications in the fields of chemistry, physics, biochemistry, and materials science, and his papers have been cited over 80,000 times (h-index=98).<\/p>\n
Prof. Raghavachari is an elected member of the \u201cInternational Academy of Quantum Molecular Science\u201d, and is a Fellow of the American Physical Society, the Royal Society of Chemistry, and the American Association for the Advancement of Science. He has won many prestigious awards including the \u201cDavisson-Germer Prize in Surface Physics\u201d sponsored by the American Physical Society (2009), \u201cDistinguished Alumnus\u201d award at Indian Institute of Technology, Madras, (2014), Chemical Research Society of India Medal (2017), the Bicentennial Medal for Distinguished and Distinctive Service, Indiana University (2020), and most recently the American Chemical Society Award for \u201cComputers in Chemical and Pharmaceutical Research\u201d.<\/p><\/div>\n\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div>
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
Abstract<\/h1><\/div>\n\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
The failure of available density functional (DFT) methods to compute accurate electronic structures and energies for complex systems is well documented. While newly developed dispersion corrected and long-range corrected density functionals offer improved results, the errors in thermochemical properties computed with DFT are often greater than 5 kcal\/mol for many chemical processes. In this talk, we discuss a new fragmentation-based molecular representation framework designed for delta machine learning (\u2000-ML) with the central goal of correcting the deficiencies of approximate methods such as DFT to achieve a high level of accuracy that is currently only achievable with sophisticated and computationally intensive methods such as coupled cluster theory.<\/div>\n\t\t\t<\/div>\n\t\t\t<\/div>
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
Our framework is based on a judicious combination of ideas from fragmentation, error cancellation, and a state-of-the-art deep learning architecture. Broadly, we develop a general interpretable framework for molecular machine learning by incorporating the inherent advantages prebuilt into existing error cancellation methods such as the generalized Connectivity-Based Hierarchy. More specifically, we develop an attributed graph representation through a fragmentation-based encoding of local chemical environments. In addition, we incorporate electronic descriptors from DFT calculations in the \u2000-ML model to achieve high accuracy for the calculation of electronic properties. Our QM\/ML framework and its performance for theoretical thermochemistry, redox potentials, pKa calculations, and NMR chemical shifts are analyzed and assessed in this presentation.<\/div>\n\t\t\t<\/div>\n\t\t\t<\/div>
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t
![\"Krishnan_Raghavachari_graphics\"](\"https:\/\/events.mcs.cmu.edu\/pople2024\/wp-content\/uploads\/sites\/36\/2024\/08\/Krishnan_Raghavachari_graphics_CMU-1.png\" "\\"Krishnan_Raghavachari_graphics\\"")
<\/span>\n\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div><\/p>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":17,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"class_list":["post-770","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/pages\/770","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/users\/17"}],"replies":[{"embeddable":true,"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/comments?post=770"}],"version-history":[{"count":31,"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/pages\/770\/revisions"}],"predecessor-version":[{"id":890,"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/pages\/770\/revisions\/890"}],"wp:attachment":[{"href":"https:\/\/events.mcs.cmu.edu\/pople\/wp-json\/wp\/v2\/media?parent=770"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"John](\"https:\/\/events.mcs.cmu.edu\/pople2024\/wp-content\/uploads\/sites\/36\/2024\/08\/pople_lecture_hero-1.jpg\" "\\"John")
\n\t\t\t\n\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t<\/div>![\"Krishnan_Raghavachari\"](\"https:\/\/events.mcs.cmu.edu\/pople2024\/wp-content\/uploads\/sites\/36\/2024\/08\/Krishnan_Raghavachari_photo_sq.jpg\" "\\"Krishnan_Raghavachari\\"")
<\/span>\n\t\t\t<\/div>\n\t\t\t<\/div>