The rising demand for sustainable solutions to technological and societal challenges has driven significant research and development efforts in computational chemistry and materials science. Indeed, the need to investigate the properties of large and complex systems (e.g., drugs, proteins, (in)organic semiconductors) has motivated marked advancements in quantum-mechanical (QM) methods as well as their integration with machine learning (ML) techniques to develop data-driven frameworks for accelerating diverse academic and technological projects. Despite these substantial advancements in recent years, challenges remain in developing Efficient, Accurate, Scalable, and Transferable (EAST) methodologies that minimize energy consumption and data storage when creating robust ML models. Achieving EAST requirements will contribute to the sustainable exploration of the chemical space encompassing molecules and materials. To find novel alternatives to address this issue, the SusML workshop will focus on discussing the advantages and shortcomings of two relevant topics:
--Data-efficient ML-based computational methods
--Inverse property-to-structure problem
The primary objective of the SusML workshop is to foster dynamic discussions on these topics, aiming to generate innovative ideas for the systematic coupling of advanced ML techniques with QM methods to generate EAST methodologies---a critical element for sustainable exploration (both directly and inversely) of the chemical space encompassing molecules and materials. We envision our workshop as a pivotal milestone capable of sparking new collaborative efforts aimed at further developing ML-aided investigations that can have profound implications for accelerating chemical and material research, streamlining compound discovery, and cultivating environmentally friendly practices.
The rising demand for sustainable solutions to technological and societal challenges has driven significant research and development efforts in computational chemistry and materials science. Indeed, the need to investigate the properties of large and complex systems (e.g., drugs, proteins, (in)organic semiconductors) has motivated marked advancements in quantum-mechanical (QM) methods as well as their integration with machine learning (ML) techniques to develop data-driven frameworks for accelerating diverse academic and technological projects. Despite these substantial advancements in recent years, challenges remain in developing Efficient, Accurate, Scalable, and Transferable (EAST) methodologies that minimize energy consumption and data storage when creating robust ML models. Achieving EAST requirements will contribute to the sustainable exploration of the chemical space encompassing molecules and materials. To find novel alternatives to address this issue, the SusML workshop will focus on discussing the advantages and shortcomings of two relevant topics:
--Data-efficient ML-based computational methods
--Inverse property-to-structure problem
The primary objective of the SusML workshop is to foster dynamic discussions on these topics, aiming to generate innovative ideas for the systematic coupling of advanced ML techniques with QM methods to generate EAST methodologies---a critical element for sustainable exploration (both directly and inversely) of the chemical space encompassing molecules and materials. We envision our workshop as a pivotal milestone capable of sparking new collaborative efforts aimed at further developing ML-aided investigations that can have profound implications for accelerating chemical and material research, streamlining compound discovery, and cultivating environmentally friendly practices.
