Singapore

Many industries use deep learning algorithms to assist in the research and development of products. Therefore, we focus on using deep learning to help predict the parameters of new compounds and to help determine whether the compound is stable.
In this project, we explore the establishment of AI models through different algorithms for machine learning and deep learning. Then, we obtain the important features in the prediction process from the machine learning part, so as to provide a reference for actual research work. The accuracy of the model we get is analyzed and compared to determine the final algorithm and build the model. Using this model, we can predict the formation energy and energy above hull of compounds to determine whether they can exist stably. Finally, we integrate functions into a GUI, so that researchers can use this GUI to make stable predictions of any unknown new compounds easily and conveniently. Throughout the project, the accuracy of the model and the GUI have been continuously improved by us. And show these results in the following part.

MATERIALS FOR HUMANITY (MH 21)

Deep learning for accurate predictions of alloy stability

[![](https://assets.underline.io/uploads/markdown_image/1/image/9a88ae95fbde06e80f4bcf6d4f7d133a.png)](https://certificateaixia.paperform.co/)
The certificate will be provided by December 17. 
In case you need it urgently please contact: francesca.gasparini@unimib.it. 

 

The 20th International Conference of the Italian Association for Artificial Intelligence (**AIxIA 2021**) is organized by AIxIA (Associazione Italiana per l’Intelligenza Artificiale), which is a non-profit scientific society founded in 1988 and devoted to the promotion of Artificial Intelligence. The society aims to increase the public awareness of AI, encourage the teaching of it and promote research in the field.

 
 



**General Chair**

Stefania Bandini, University of Milano-Bicocca, Italyand RCAST - The University of Tokyo, Japan


**Program Chairs**

Viviana Mascardi, University of Genova, Italy; Matteo Palmonari, University of Milano-Bicocca, Italy; Giuseppe Vizzari, University of Milano-Bicocca, Italy

**Organizing Committee**

Francesca Gasparini, University of Milano-Bicocca, Italy (**Chair**); Stefano Ferilli, University of Bari, Italy; Marta Giltri, University of Milano-Bicocca, Italy; Daniela Briola, University of Insubria, Italy


**Publicity and Communication Committee**

Elisabetta Fersini, University of Milano-Bicocca, Italy (**Chair**); Angelo Ferrando, University of Genova, Italy; Francesca Alessandra Lisi, University of Bari, Italy; Giovanni Stilo, University of L'Aquila, Italy; Marco Viviani, University of Milano-Bicocca, Italy


**Workshop Chairs**

Gabriella Cortellessa, ISTC-CNR, Rome, Italy; Silvia Rossi, University of Naples Federico II, Italy; Fabio Stella, University of Milano-Bicocca, Italy


**PC Members**

Fulvio Mastrogiovanni, University of Genova, Italy; Roberto Basili, University of Rome Tor Vergata, Italy; Nicola Gatti, Polytechnic University of Milan, Italy; Stefano Bistarelli, University of Perugia, Italy; Piero Poccianti, Consorzio Operativo Gruppo MPS, Italy; Francesco Scarcello, University of Calabria, Italy; Flavio S. Correa Da Silva, University of São Paulo, Brazil; Elena Cabrio, Université Côte d'Azur, CNRS, Inria, I3S, France; Antonio Lieto, University of Turin, Italy; Diego Calvanese, Free University of Bozen-Bolzano, Italy; Alessio Micheli, University of Pisa, Italy; Francesca A. Lisi University of Bari Aldo Moro, Italy; Mauro Vallati, University of Huddersfield, United Kingdom; Bernardo Magnini, Fondazione Bruno Kessler (FBK), Italy; Luigia C. Aiello, Sapienza University of Rome, Italy; Rafael H. Bordini, Pontifical Catholic University of Rio Grande do Sul, Brazil; Chiara Ghidini, Fondazione Bruno Kessler (FBK), Italy; Federico Bergenti, University of Parma, Italy; Francesco Ricca, University of Calabria, Italy; Andrea Burattin, Technical University of Denmark, Denmark; Angelo Oddi, ISTC-CNR, Italian National Research Council, Italy; Farshad Badie, Aalborg University, Denmark; Emanuel Sallinger, University of Oxford, United Kingdom; Marin Lujak, University Rey Juan Carlos, Spain; Giancarlo Fortino, University of Calabria, Italy; Salvatore Gaglio, University of Palermo, Italy; Johannes K. Fichte, TU Dresden, Germany; Matteo Baldoni, University of Turin, Italy; Fabrizio Riguzzi, Univeristy of Ferrara, Italy; Ivana Dusparic, Trinity College Dublin, Ireland; Serena Villata, CNRS - Laboratoire d'Informatique, Signaux et Systèmes de Sophia-Antipolis, France; Paolo Rosso, Technical University of Valencia, Spain; Ana L. C. Bazzan, Federal University of Rio Grande do Sul, Brazil; Gianluigi Greco, University of Calabria, Italy; Stefano Cagnoni, University of Parma, Italy; Gian Luca Pozzato, University of Turin, Italy; Salvatore Ruggieri, University of Turin, Italy; Claudio Di Ciccio, Sapienza University of Rome, Italy; Eloisa Vargiu, Cetaqua, Water Technology Center, Spain; Marco Maratea, University of Genova, Italy; Andrea Roli, University of Bologna, Italy; Viviana Patti, University of Turin, Italy; Stefania Costantini, University of L'Aquila, Italy; Antonius Weinzierl, Vienna University of Technology, Austria; Silvia Rossi, University of Naples Federico II, Italy; Giovanni Semeraro, University of Bari, Italy; Luciano Serafini, Fondazione Bruno Kessler (FBK), Italy; Laura Giordano, University of Eastern Piedmont, Italy; Ernesto Jimenez-Ruiz, City, University of London, United Kingdom; Marco Gavanelli, University of Ferrara, Italy; Federico Chesani, University of Bologna, Italy

Please register with AIxIA to be able to enter the conference on Underline.io. If you have already registered, please login to Underline.io to be able to see full conference content. In case you do not have an Underline account, please create one. Thank you

Please login to your Underline account

AIxIA 2021

The 20th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2021) is organized by AIxIA (Associazione Italiana per l’Intelligenza Artificiale), which is a non-profit scientific society founded in 1988 and devoted to the promotion of Artificial Intelligence. The society aims to increase the public awareness of AI, encourage the teaching of it and promote research in the field.

technical paper

**Please view our Welcome to MH21 video.**

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1UYHV1pcz?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>


**Walkthrough Video**

Here is a Walkthrough showing how to navigate the conference and network with others 

<center>
<figure class="video_container">
	 <iframe src="https://screencast-o-matic.com/watch/cr1vnrV1paE?v=6&ff=1&title=0&controls=1" width=826 height=526 frameborder="0" allowfullscreen="true"> </iframe>
</figure>
	</center>




	 Organized by
 
<a href="https://mrs.org.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/95d5789689d5f900b90e9fe54ea3cf0d.jpg"</a>

 
	
	 In association with
 
<a href="http://www.ntu.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/334cd171ad2fdee033c635b368e5e160.png"</a>
	<a href="https://www.nus.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/32e6f82c0440f7bb24c590f9f67821bd.png"</a>
		<a href="https://www.a-star.edu.sg/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/72557865b589cfa3d2a0243ffce4a270.png"</a>
	
 


	 Supported by
 
		 <a href="https://www.visitsingapore.com/mice/en/about-us/about-secb/">
<img src="https://assets.underline.io/uploads/markdown_image/1/image/6e679842ead62be522ba6a630b82d5c9.png"</a>

To access the MH 21 Conference, [please register](https://mh21.mrs.org.sg/event/da3f9651-429f-4763-81b2-ebd7311b6bc8/websitePage:206f7605-1731-41f9-8421-56d3890f398f)

MH 21 is arranged around 6 themes that focus attention on pressing needs that should be addressed to enhance the well-being of communities and provide a forum for presenting breakthrough discoveries.

The setting up of the magnetic topological materials database provides the possibility of exploring in this field through the machine learning method. Here, we trained a support vector classification (SVC) model to predict the topological state of the magnetic materials. There are 551 of materials in the topological magnetic materials database, and for each material, the topological state is different for different U values as the date are the DFT results. So, different U values should be different data and finally, 1319 of data are used for the model training. Although the strongly imbalance of the data (only 87 of data are topological insulators (TIs)), the trained model still can reach high precision (0.99) for trivial materials, which means that almost all the un-trivial magnetic materials can be screened by our model. Then, the SVC model is used to predict the topological state of the magnetic materials in MAGNDATA database and select 65 materials as TIs and 15 materials as enforced semimetals (ESs) for all the four times prediction. Topological quantum chemistry method then used to test the topological state of the randomly selected 11 of the un-trivial materials and indicate that 5 of them are ES and 1 is TI. Then first principle method is applied to further inspect the topological state of the only TI. The atom unfilled orbitals and valence electrons are deemed as import features by our model, which is in consistence with the physical analysis results. This SVC model could help screen topological materials from large amount of magnetic materials.

Machine Learning Method to Fastly Screen Topological Magnetic Materials

Two-dimensional topological superconductivity candidate in a van der Waals layered material
Jing-Yang You ,1 Bo Gu ,2 Gang Su,2 and Yuan Ping Feng1,3,*
1Department of Physics, National University of Singapore, 2 Science Drive 3, Singapore 117551
2Kavli Institute for Theoretical Sciences, and CAS Center for Excellence in Topological Quantum Computation, University of Chinese Academy of Sciences, Beijing 100190, China
3Centre for Advanced 2D Materials, National University of Singapore, 6 Science Drive 2, Singapore 117546

Two-dimensional (2D) topological superconductors are highly desired because they not only offer opportunities for exploring novel exotic quantum physics but also possess potential applications in quantum computation. However, there are few reports about 2D superconductors, let alone topological superconductors. Here, we find a 2D monolayer W2N3, which can be exfoliated from its real van der Waals bulk material with much lower exfoliation energy than MoS2, to be a topological metal with exotic topological states at different energy levels. Owing to the Van Hove singularities, the density of states near the Fermi level are high, making the monolayer a compensate metal. Moreover, the monolayer W2N3 is unveiled to be a superconductor with the superconducting transition temperature TC∼22 K and a superconducting gap of about 5 meV based on the anisotropic Migdal-Eliashberg formalism, arising from the strong electron-phonon coupling around the point, and the 2D superconductor is phonon mediated and fits the BCS mechanism with an Ising-type pairing. Because of the strong electron and lattice coupling, the monolayer displays a non-Fermi liquid behavior in its normal states at temperatures lower than 80 K, where the specific heat exhibits T3 behavior and the Wiedemann-Franz law is dramatically violated. Our findings not only provide a platform to study the emergent phenomena in 2D topological superconductors, but also open a door to discover more 2D high-temperature topological superconductors in van der Waals materials.

Reference:
[1] J-Y. You, B Gu, G. Su, Yuan Ping Feng, Physical Review B 103, 104503 (2021).

Two-dimensional topological superconductivity candidate in a van der Waals layered material

Two-dimensional (2D) Janus MoSSe have great potential for photocatalytic water splitting due to their novel properties induced by the unique out-of-plane asymmetric structures.1-3 Doping is an effective way to modulating the photocatalytic properties. The intrinsic defects are usually unavoidable when introducing extrinsic defects like transitional metal adatoms. However, to our knowledge, few researchers have investigated the synergistic effects on the photocatalytic properties when vacancies and adsorbed transition metals coexist on the MoSSe surface. Here, we systematically investigate the geometric, electronic, and optical properties of 2D Janus MoSSe with titanium doping and vacancies to explore their synergistic effects on photocatalytic activity. 
We find that there is effective attraction between the substituted or adsorbed Ti atoms and S/Se vacancies. The Ti adatoms dramatically extend the light absorption range to the infrared region. The S/Se vacancies coexisting with Ti adatoms will modulate the transition of photoexcited electrons, thereby enhancing sunlight absorption as shown in Figure 1. The Ti adatoms either existing alone or coexisting with vacancies introduce smaller lattice distortion compared with substituted Ti atoms, and these Ti adatoms induce smaller effective mass of charge carriers. The configuration of S vacancy coexisting with Ti adatoms on the Se surface exhibits the most significant synergistic effects and the best overall photocatalytic performance. 
Our work reveals the mechanisms and effects induced by doping and vacancies coexisting in 2D Janus MoSSe and also proposes a new practical strategy to improve the performance of 2D photocatalysts.


Effects of Vacancy and Ti Doping in 2D Janus MoSSe on Photocatalysis

We present a quantum-mechanical study of a B2-phase of Fe70Al30 alloy
with and without antiphase boundaries (APBs) with the {001} crystallographic
orientation of APB interfaces. Our motivation is to explain experimental
findings demonstrating a significantly higher magnetic flux density from 
A2-phase interlayers at the thermally-induced APBs in this material
and suggesting that the ferromagnetism is stabilized by the disorder in
the A2 phase. Our investigation of sharp APBs (without any A2-phase interlayer) 
indicates that they have moderate APB energies (~0.1 J/m2) and cannot explain the 
experimentally detected increase in the ferromagnetism because they 
often induce a ferro-to-ferrimagnetic transition. When studying
thermal APBs, we introduce a few atomic layers of A2 phase of Fe70Al30 into
the interface of sharp APBs. The averaged computed magnetic moment of Fe atoms
in the whole B2/A2 nanocomposite is then increased by 11.5% w.r.t. the B2 phase.
The A2 phase itself (treated separately as a bulk) has the total magnetic moment
even higher, by 17.5%, and this increase also applies if the A2 phase at APBs
is sufficiently thick (the experimental value is 2–3 nm). We link the changes
in the magnetism to the facts that (i) the Al atoms in the first nearest
neighbor (1NN) shell of Fe atoms nonlinearly reduce their magnetic moments
and (ii) there are on average less Al atoms in the 1NN shell of Fe atoms
in the A2 phase. These effects synergically combine with the influence
of APBs which provide local atomic configurations not existing in
an APB-free bulk. The identified mechanism of increasing the magnetic
properties by introducing APBs with disordered phases can be used as
a new designing principle when developing new magnetic materials.

Enhancing magnetism in Fe-Al alloys by increasing chemical disorder

Solid -state conductors have attracted wide attention due to their benefit of enhanced safety and capacity with application in batteries. Our computational tools--bond valence pathway analyzer (BVPA) have been reported to provide an automated pathway analysis method for large database of solid Ionics and a dopant predictor to easily monitor the site substitution of the lattice[1]. In this work, local coordination environment characterization and classification has been introduced to further understand the interstitial site and atomic sites in the crystal structure and to improve the predicted stability of modified lattice. The local coordination number evaluation is implemented based bond valence theory [2] with proofed credibility after benchmarked with most reported methods[3]. Over 35000 atomic sites have been analysed from over 7000 high quality structure obtained from ICSD database. Machine learning has been applied to divide the local environment into over 60 defined polyhedral environments. The polyhedral environments are further explored to understand the interstitial sites in ionic pathway.

Coordination environment analysis for interstitial sites in solid-state ionic conductors.

Grain boundaries (GBs) widely exist in two-dimensional (2D) and three-dimensional (3D) materials in experiment, which significantly affect the thermoelectric performance because of the scattering effect on the transport of both electron and phonon. Motivated by the research progress in 3D SnSe [Zhao et al. Science, 2018, 360, 778], we have systematically studied the GBs in a SnSe monolayer including their stable geometric configurations, the effect of GBs on the thermoelectric properties, and the strategies to improve the performance by using first principles calculations combined with semiclassical Boltzmann theory. We find that the GBs increase the potential energy barrier of carriers and decrease the valley degeneracy of the conducting bands, leading to the reduction of Seebeck coefficient, as compared to that of the pristine SnSe sheet. We further demonstrate that, the trapping gap states are effectively eliminated or reduced by doping germanium or silicon, leading to the enhanced electrical conductivity and reduced thermal conductivity, thus the thermoelectric performance of the SnSe sheet with GBs is synergistically enhanced. These findings shed lights on developing practical strategies for modulating the thermoelectric performance of 2D polycrystalline sheets.

The effect of grain boundary on the thermoelectric performance of 2D materials: a case study of monolayer SnSe

Thermoelectric materials convert heat to electric power, and their performance quantified by the figure of merit, zT = (α2σ/κ)T. The data-driven screening combined with high-throughput calculations efficiently helps to find novel thermoelectric materials. The transport descriptors are one of the key factors to discover efficient thermoelectrics from the materials database. In our study, using a data-driven screening, we selected few potential candidates with trigonal ABX2 family and calculate the charge transport properties from first principles. Our results suggest that carrier scattering by ionized impurities and polar optical phonons are dominate in the electronic transport contrary to the often assumed acoustic-phonon-dominated scattering. We have derived new transport descriptors for the carrier mobility and the thermoelectric power factor. And along with thermal conductivity calculations using Peierls and interbranch coherent contributions, we conclude that the selected trigonal ABX2 family as potential thermoelectrics.

Data Driven Thermoelectrics: Electronic transport descriptors

Transition metal dichalcogenides exhibit polymorphs, such as 2H (trigonal coordination) and 1T (octahedral coordination) phases, which tends to have semiconducting and metallic nature, respectively. The distorted octahedral structures (Metastable 1T′-phase) show semi-metallic nature have immense interest for various applications in spintronics, energy harvesting and quantum devices etc. However, high-quality metastable 1T′-TMD synthesis without intercalation and external bias remains a great challenge. We have reported a general synthetic method for the large-scale preparation of metastable 1T′-phase TMDs and revealed the existence of superconductivity at TC = 8.6 K without any external gate bias, intercalation and pressure in 1T′-WS2. And have shown the thickness dependent and 2D-superconductivity features in our 1T′-WS2.

Low temperature electrical transport of metastable 1T′-WS2

Metal additive manufacturing is an invaluable part of robust and on-demand supply chains of the future. With its ability to consolidate material to high density into geometrically complex structures, selective laser melting (SLM) is a powerful tool for the fabrication of high strength, lightweight components. A limited pool of qualified alloys remains to be an impediment to greater adoption. Martensitic steels are a class of alloys that see widespread use in industries for tooling and highly stressed load bearing parts. However, their presence in the literature remains sparse in comparison to several workhorse steel alloys. In this work, we discuss the print optimization for this class of steels, including optimization of laser power, scanning speed, hatch distance and platform heating temperature. We investigate the general microstructural features of these steels, which is dominated by post-solidification displacive martensitic transformation and its response to cyclic thermal cycles in SLM. We demonstrate that martensitic high strength steels can be processed by SLM to have excellent mechanical properties in the as-printed state. Greater understanding of the SLM processing of this class of steels will enable the fabrication of stronger and lighter parts for increasingly demanding applications.

Untapped strength in metal additive manufacturing: characteristics and properties of martensitic steels by selective laser melting

316L stainless steel is being extensively employed as structural material in infrastructural, automotive, aerospace, and nuclear energy applications. However, one of the major drawbacks which limits its wider applications is its low yield strength (170-300 MPa in annealed state). We explored the strengthening of 316L stainless steel with dispersed ceramic particles through selective laser melting (SLM). The feedstock, powder mixture of 316L powder and ceramic particles, was mixed using low energy ball milling process. This method has a neglectable change on powder’s flowability and shape. Results showed that the ceramic particles were uniformly dispersed in the 316L matrix after SLM. In the meantime, the grain size of the 316L was significantly refined. Tensile tests showed that the yield strength of the strengthened 316L was increased about 40% (from 599 to 832 MPa). Meanwhile, the 316L stainless steel with dispersed ceramic particles has a good combination of high strength and ductility. The strength enhancement is primarily attributed to grain refinement and Orowan strengthening. These findings offer a new perspective on the strengthening of 316L stainless steel.

Premium content

Deep learning for accurate predictions of alloy stability

Downloads

Next from MATERIALS FOR HUMANITY (MH 21)

Machine Learning Method to Fastly Screen Topological Magnetic Materials

Similar lecture

Fairness and Popularity Bias in Recommender Systems: an Empirical Evaluation