State of the Art and Workshop Objectives
Data-driven methods have emerged as a novel paradigm to advance materials discovery over the past decade. Machine learning potentials (MLPs) enable the sampling of trajectories with the same accuracy of high-level electronic structure methods but at a fraction of their cost. MLPs have been established as a means to rationalize puzzles previously unapproachable by atomistic simulations. Elsewhere, the chemical and physical properties of large chemical spaces are now screened in a high-throughput fashion by leveraging artificial intelligence methods, materials simulations, and automation protocols. The screening is not only viable for the case of known structures, but generative models can now autonomously generate previously-unseen, and tailored, molecules and crystals structures with a target property. Machine learning (ML) methods therefore serve as formidable surrogates to accelerate expensive computational screening, but also to guide experimental screening and extract knowledge from data gathered via high-throughput or from literature. Furthermore, the advances in the theoretical understanding of how machine learning algorithms work is demystifying and surpassing the vision of data-driven approaches as magic black-boxes.
This event built upon the state-of-the-art in the field of machine learning for materials in two ways. Firstly, it helped instruct the next generation of young researchers on the latest advancements in methods and applications of AI for material discovery through didactic lectures and hands-on tutorials. Secondly, the workshop promoted a discussion on the implications of the latest advancements in data-driven methods on the different sub-areas of Materials discovery, bringing together experts of different fields in the world of machine learning for materials and promoting cross-contamination of ideas and techniques.
The introductory part of the workshop paved the way to an overview of supervised and unsupervised methods which well represented the state of the art in machine learning for materials. A number of tutorials on publicly available open source and documented codes were offered to participants.
Many discussions during the workshop focused on the design of atomic descriptors for supervised tasks in materials science. Two techniques were highlighted by various speakers as top performers: atom-density (e.g., atomic cluster expansion (ACE)) representation, and equivariant learnt representations via message-passing networks (MPE(3)N). Both methods efficiently encode information about local atomic environments and allow for very accurate learning of atomic or structural properties (e.g., forces, energies, etc.). A discussion of a unified theory to reconcile the dichotomy between ACE and MPE(3)N was a recurring trend across invited and contributed speakers. State-of-the-art Pareto fronts of prediction speed-accuracy, and an analysis on the memory and time requirements for training were also often reported.
A second set of common themes and techniques related to the use of generative models. Their application ranged across disciplines: SMILES-based short-term memory recurrent neural networks were used to design drugs; cartesian coordinates and autoencoders were utilized to unbiasedly obtain equivariant representation for quantitative structure-activity relationships; classical descriptors and variational autoencoders were adopted to map states during dynamics and/or glass phenotypes.
A third set of common themes related to the use of machine learning methods to accelerate the first principles screening of material stability. The application of this method ranges from energy materials (e.g., perovskites) to molecular crystals (e.g., drugs) and leveraged uncertainty-driven methods to iteratively and accurately chart convex-hulls and establish thermodynamically stable phases.
A final class of major scientific points of discussion encompasses a broader spectrum of topics which enables to bridge complexity gaps between data models and experiments, as so to establish rational design paradigms from structure-property relationships. A heterogeneous list of techniques debated includes (but is not limited to): machine learning potentials for fast-and-accurate simulation of complex dynamics; transfer learning to derive universal predictors which work well across different chemistries; experimental characterization and manipulation of materials via data-driven optimization.
Overall, the need for cross-contamination of expertises emerged as a strong and resonant topic throughout the conference. During panel discussions, presentations, and face-to-face interactions, participants expressed the need to escape scientific bubbles and gather information about techniques, applications, and developments in fields adjacent to their own research. In this regard the presence of leaders in atomistic modeling, computer science, machine learning, experimentalists, and industry representatives enabled an interdisciplinary exchange of perspectives and experiences.
Our workshop was indeed specifically designed to address the needs for multi-disciplinary cross-contamination, and we received resounding feedback about how such an effort was successful. All (to our knowledge, and according to a currently ongoing survey) participants to the workshop, be it an invited speaker, an online attendee, a poster presenter, or a young researcher that attended their first conference on the topic, was largely positive about the structure, topics, and organization of the event.
Finally, all talks, tutorials, and panel discussions that took place during the workshop have been recorded and uploaded on Youtube, the ICTP website, and the conference’s website, thus making high-quality scientific content available to anyone.
The workshop allowed researchers in adjacent fields to meet each other, learn about the most recent advances of their colleagues, and network in a scientifically fertile environment. Moreover, the presence of an introductory school in the workshop allowed for young researchers that are new to the field to learn about potential applications of machine learning technologies to their area of interest, and to better appreciate the advancements presented by invited leading researchers during the “workshop” part of the conference. From the networking perspective, the event north star was indeed to enable the cross-fertilization of research networks, promoting the encounter and collaboration between domain experts.
From the computational perspective, tutorials used Google Colabs seamlessly. All the codes discussed were open-source, and all tutorials presented during the workshop are available on the conference’s website for anyone to follow.
The computational expense associated with the machine learning for materials codes described is mostly related to data-generation. In this case data from the literature were utilized. A discussion on how to develop accessible and efficient machine learning codes which do not necessarily necessitate expensive computational architectures (e.g., GPU highly parallel facilities) has been put forward. Similarly, a reflection on the need to push open-science (open code, open data, etc.) to ensure the democratization of the field has been discussed. While there exist open repositories and robust generation routine for computational data (e.g., Materials Cloud, IoChemBD, NOMAD, Materials Project, AFLOW ), a discussion on how to promote the creation of FAIR compliant routines and databases for experiment-related data and codes was initiated
Scientific, Technologic, and Societal Impact
The discovery of novel materials for catalytic applications, energy storage, diagnostics, and therapeutics is one of the key ways in which the goal of a sustainable and equitable development can be reached (see also UN Sustainable Development Goals (SDGs) )
The recent years have seen a surge in the development and application of machine learning technologies in materials science. The initial results are extremely promising, with applications ranging from the design of novel catalysts for CO2 reduction to the exploration of the chemical space of energy materials for novel Lithium-free batteries. While giant steps have been made within the design of algorithms and the understanding of the theoretical backbone of machine learning in materials and chemical science, widespread and large scale applications are just starting to bloom and to have a real-world impact, allowing, e.g., for the discovery of novel stable materials or the design of never-seen-before catalysts or drugs.
This workshop pushed forward research in these critical fields by providing both a way to spread research advancements to young researchers, and a way to initiate collaboration between widely renowned scientists of different fields. This workshop further equipped the next generation of scientists (in academia and industry alike) with skills in tackling the complex problems related to high-performance materials design.
The presence of industry representatives (Roche, Bayer, Microsoft Research, AIndo) on one hand offered an overview of possible career pathways to participants. From an alternative perspective, the state-of-the-art methods and achievements obtained by our community were promoted to these R&D teams.
Molecular Simulation 2022: past, present and future
Event website: https://bricabrac.fisica.unimo.it/ErcMlk80/
The meeting took place on 25th to 29th June 2022 in Erice (Italy), at the Villa San Giovanni complex, previously a clerical summer residence but now used for conferences and cultural manifestations.
This event brought together old and new friends to discuss state of the art methods and current challenges in molecular simulations, reflect on many years of development and applications, and reflect on the future of the field. The meeting enabled scientists from different generations to meet in an atmosphere that combined excellence and open discussions and paved the way for new scientific perspectives and collaborations. There were 29 speakers coming from all over the world, and over 125 participants in total (full capacity of Erice site that had some restrictions in place due to the Covid pandemic). The program was composed of 9 sessions each with 3 to 4 speakers, and there were several poster sessions. The meeting was also an occasion to celebrate Prof Mike Klein’s 80th birthday, the numerous important and remarkable contributions Mike has made to chemistry, biophysics, materials science, and, in particular, the field of computer simulation. Speakers highlighted in their talks recent advances in modelling and simulation in biophysics, biochemistry, material science, chemistry and physics.
Eight PhD students were awarded prizes among the 64 posters presented in the meeting. The prizes were contributed by the MDPI publisher, CECAM and the RSC.
We thank Psi-k for their generous support in making this a successful meeting.
CECAM/Psi-K Flagship Workshop “Light-matter interaction and ultrafast nonequilibrium dynamics in plasmonic materials”
July 18, 2022 – July 21, 2022, University of Warwick, UK
Description of Event
The Psi-K & CECAM sponsored meeting “Light-matter interaction and ultrafast nonequilibrium dynamics in plasmonic materials” was held from 18th to 21st of July 2022 at the University of Warwick. It featured 28 talks, 4 discussion sessions, and 10 posters. It was attended by 42 in-person attendees from 12 different countries and broadcast as a webinar with between 3 and 17 virtual attendees at any time.
A full theoretical description of light-matter interaction and plasmon-induced ultrafast non-equilibrium dynamics is a formidable challenge that demands an intrinsically multidisciplinary and multiscale approach. A variety of different approaches based on time-dependent Density Functional Theory, many-body perturbation theory, molecular dynamics, Mie theory, continuum electrodynamics, and combinations thereof have emerged in recent years to address many of the open questions in plasmonics. Further improvements in theoretical descriptions are crucial to optimize SPP generation and amplification in materials, to tailor losses and plasmonic lifetimes, as well as to integrate plasmonic effects into semiconductor technology to create new quantum materials. Due to the diverse aspects of this problem, a coherent research community around theoretical plasmonics is only slowly emerging.
The aim of this workshop was to assess the state of computational methods in this field, to identify major challenges, as well as to provide engagement between disparate communities to create space for cross-community collaboration. Continue reading Psi-K/CECAM Flagship workshop “Light-matter interaction and ultrafast nonequilibrium dynamics in plasmonic materials”
10th ABINIT International Developer Workshop – Part 2
May 16-19, 2022 Guidel-Plages, France
The ABINIT developer workshops form a series of events, crucial for the community of ABINIT developers, organized every two years. A unique occasion for most developers to acquire or maintain a global view of the project and stay up to date with the latest capabilities, planned developments, and overall strategy. The developer workshop is always an opportunity to invite external researchers, from other codes and communities, to exchange best practices and expertise…
This workshop was the second part of the 10th ABINIT developer workshop. It was held from 16th to 19th May 2022 in Guidel-Plages (Brittany, France).
Because of the COVID19 pandemic situation, in June 2021, the meeting occurred in a fully remote version, with only remote presentations and some group discussions. We missed several important parts of the workshop : small group discussions, thematic discussions, informal discussions, hackathons, etc. Many of the participants emphasized the need to meet again in person when the health situation permits it.
In May 2022 we organized the second part of the workshop, with a smaller number of participants, mostly based on the missing ingredients above, plus a few invited presentations. It was a complementary and entirely live/offline event, consisting of discussions, round tables and hackathons. The physical presence of developers was a requirement to have efficient round tables and informal discussions.
The workshop was mainly dedicated to implementations and decision making by the developers:
Every morning we had a session of hackathons. Divided into small groups, we worked on the ABINIT package : coding, improving the documentation, creating tutorials, interfacing the code with other software, etc. Each developer chose projects and hackathons according to his/her specific expertise in the project.
During two afternoons, we met collectively to discuss and consider the future of the code: future scientific themes, dissemination and the visibility, user experience improvement.
A third afternoon was dedicated to invited speakers’ presentations. The speakers were chosen because of their involvement in projects external/complementary to ABINIT. Continue reading 10th ABINIT International Developer Workshop – Part 2
From 20th until 24th June 2022 we organised a workshop on the theme of “Error control in first-principles modelling” at the CECAM Headquarters in Lausanne (workshop website). For one week the workshop unified like-minded researchers from a range of communities, including quantum chemistry, materials sciences, scientific computing and mathematics to jointly discuss the determination of errors in atomistic modelling. The main goal was to obtain a cross-community overview of ongoing work and to establish new links between the disciplines.
Amongst others we discussed topics such as: the determination of errors in observables, which are the result of long molecular dynamics simulations, the reliability and efficiency of numerical procedures and how to go beyond benchmarking or convergence studies via a rigorous mathematical understanding of errors. We further explored interactions with the field of uncertainty quantification to link numerical and modeling errors in electronic structure calculations or to understand error propagation in interatomic potentials via statistical inference.
Scientific Report for the workshop “Astrochemistry meets Surface Science: Theoretical Frontiers”
April 5th – 8th 2022
Aarhus Institute of Advanced Studies, Aarhus University, Denmark
The goal of the workshop was to bring closer together the research communities of theoretical Astrochemistry and theoretical (under Earth conditions) Surface Science. These two fields often address very similar questions, while using the exact same techniques and methodologies (e.g. electronic structure methods such as density functional theory, molecular dynamics or kinetic Monte Carlo simulations). And yet, despite these similarities, a noticeable communication gap exists between the two communities. Our multidisciplinary workshop aimed to bridge this gap and help establish new networking and collaboration ties between these fields.
The workshop covered a broad range of topics ranging from surface reaction networks and kinetic models to the characterization of interstellar ices and questions of energy dissipation and heat transport. In parallel to the underlying scientific questions, special focus was placed on theoretical and methodological aspects, as well as computational and numerical tools that are used in either one or both of the fields of astrochemistry and Earth-related surface science.
Many congratulations to the former Psi-k Chair, and current Trustee, Nicola Marzari and his team on winning the inaugural PRACE HPC Excellence Award for their work in the discovery and characterization of novel two-dimensional materials!
ICTP, Trieste (Italy), 16-27 May 2022
The last two decades have witnessed a tremendous growth in the use of Wannier functions (WFs) for first-principles electronic structure calculations. Beyond providing fundamental insights on several aspects of the electronic structure, from chemical bonding to electrical polarisation, topological invariants, Berry curvature and more, WFs have found applications in a plethora of different domains.
The software package WANNIER90 has become a reference for calculating maximally-localised Wannier functions (MLWFs) [1,2] and related properties [2,3,4]. As Wannier functions are independent from the basis sets used to represent the electronic structure in the underlying first-principles calculations, WANNIER90 can be interfaced to virtually any electronic-structure code. Indeed, most of the major electronic-structure packages have already an interface to WANNIER90, including Quantum ESPRESSO, ABINIT, VASP, Siesta, Wien2k, Fleur and Octopus.
The availability of a robust MLWF code that is connected to several ab-initio engines has acted as a fertiliser for the birth of independent computational efforts aimed at calculating complex materials properties by leveraging WFs. Several independent packages exploiting MLWFs and WANNIER90 exist nowadays, targeting a number of properties, from electron-phonon coupling  (EPW) to topological invariants  (Z2Pack), surface spectral densities  (WannierTools), Berry-phase related properties  (Wannier Berri), tight-binding models (PythTB, TBModels), high-throughput calculations  (AiiDA-Wannier90), dynamical mean field theory (TRIQS), just to mention a few.
Wannier 2022 has been an event that put together the community behind these symbiotic packages that form a research and software ecosystem built upon the concept of MLWFs. The workshop has served the two-fold objective of teaching several techniques enabled by Wannier functions to young researchers and fostering an integration between all the packages composing the Wannier ecosystem, contrasting fragmentation and duplication of efforts.
This workshop has been generously funded by ICTP, Psi-k, MaX, NCCR MARVEL and CECAM. The computational infrastructure to run the hands-on tutorials has been provided by ICTP through their ICTP Cloud.
The workshop was run by 6 directors:
- Antimo Marrazzo, University of Trieste
- Roxana Margine, Binghamton University
- Sinisa Coh, University of California Riverside
- Stepan Tsirkin, University of Zurich
- Giovanni Pizzi, EPFL
- Nicola Seriani, ICTP (local organiser)
The event comprised two parts, a summer school (first week) and a developers meeting (second week).
We are inviting young students/researchers to the 2nd edition of TREX school on QMC with TurboRVB, organised Scuola Internazionale Superiore di Studi Avanzati (SISSA) and TREX project.
The School will be held from 04-08 July 2022 in Trieste, Italy at Scuola Internazionale Superiore di Studi Avanzati (SISSA) where you will get the chance to attend keynotes presentations, lectures, and hands-on tutorials and applications coming from our HPC experts within the TREX project.
Why should you join the School?
- The TREX project is offering participants free lodging, coffee breaks, lunch breaks and a social dinner during the School. This will be the perfect occasion to meet your peers and colleagues in person, share experiences and learn more about QMC methods.
- Participants are invited to submit their poster contributions and present results and ongoing activities at the TREX School. The best poster winners will get the opportunity to be financed as visiting students for one week: one winner will be visiting the SISSA in Trieste (IT) and another winner the CNRS in Paris (FR), with food and accommodation provided by the TREX project. Please fill out the online form to submit a poster. The deadline for poster submission is 26 June 2022 at 17:00 CEST.
- Get the chance to explore beautiful Trieste, top-ranked in Italy for its quality of life, and to discover this town renowned for its scientific institutions. TREX project is not covering participant’s travel costs.
- Registration is open until 24 June 2022, 17:00 CEST
- Selected participants will be announced immediately right after the completion of each registration
- Submission for the Call for Poster is until 26 June 2022, 17:00 CEST
Interested in promoting discussion and fostering collaborations in research areas of broad mutual interest for CECAM and Psi-k?
Submit your expression of interest to organize a CECAM Psi-k Research Conference in 2023 by filling the simple form at:
Deadline for submission of the expression of interest: 4 September 2022
Diversity (specific expertise in simulation and modelling, geographical, gender, career stage…) and interdisciplinarity among organisers and participants are key evaluation criteria.
The CECAM Psi-k Research Conference is a forum to explore and foster progress in exciting new topics and open questions of interest to different communities in simulation and modeling, rather than another opportunity to showcase consolidated research. The proposed duration and format of the event should reflect this spirit, with ample time set aside for common discussions and informal exchanges.
Details of previous CECAM Psi-k Research Conferences can be found online.
If selected for further evaluation, you will be invited to submit a full proposal – format similar to previous years – at the end of September 2022, with a deadline for submission on 31 October 2022.