Update f3dasm logo and improve documentation

paper/paper.bib

@@ -1,26 +1,3 @@
-@Article{Aage2017,
-  author    = {Aage, Niels and Andreassen, Erik and Lazarov, Boyan S and Sigmund, Ole},
-  journal   = {Nature},
-  title     = {Giga-voxel computational morphogenesis for structural design},
-  year      = {2017},
-  number    = {7674},
-  pages     = {84--86},
-  volume    = {550},
-  doi       = {10.1038/nature23911},
-  publisher = {Nature Publishing Group},
-}
-
-@MastersThesis{Schelling2021,
-  author   = {Martin van der Schelling},
-  school   = {Delft University of Technology},
-  title    = {A data-driven heuristic decision strategy for data-scarce optimization: with an application towards bio-based composites},
-  year     = {2021},
-  month    = mar,
-  type     = {mathesis},
-  abstract = {Algorithmic optimization is a viable tool for solving complex materials engineering issues. In this study, a data-scarce Bayesian optimization model was developed to research the composition of bio-based composites. The proof-of-concept program adjusts the natural materials' weight ratios to optimize towards user-defined mechanical properties. Preliminary results show that the bio-composites proposed by the program had improved properties compared to existing bulk-moulding compounds. However, the algorithm choice is often arbitrary or based on anecdotal evidence. In parallel, this thesis proposed a data-driven framework for general data-scarce optimization problems to adapt the meta-heuristic during optimization. Guided by the 'No Free Lunch' theorem, we verified that the effectiveness over a selection of algorithms is dependent on problem-specific features and convergence. This effectiveness was captured in a unique identifier metric by optimizing a generated training set of optimization problems. The average solution quality was improved by combining several meta-heuristics in series, based on these problem-specifics. During the optimization of problems in the testing set, the same unique identifier was constructed at predefined stages in the optimization process. Subsequently, the problem was classified, and the meta-heuristic was adapted to the best-performing algorithm based on similar training samples. Experiments with various classifiers and a different number of predefined assessment stages were performed. Results show that the data-driven heuristic decision strategy outperformed the individual optimizers on the testing set. Despite the use of binarization techniques, the classification accuracy was heavily influenced by the imbalanced training set. In terms of computational resources, the various adaptions of the data-driven heuristic strategy are 2.5 times faster in runtime compared to the best-performing meta-heuristic Bayesian Optimization. Lastly, the framework was benchmarked against the 'learning to optimize' study and shows excellent performance on the logistic regression problem compared to the autonomous optimizer. In conclusion, it has been shown that even with the limited information of black-box optimization problems, data-driven optimization effectively improves the current standard of materials engineering processes.},
-  url      = {https://repository.tudelft.nl/islandora/object/uuid:d58271d6-21bb-470c-a5ee-4584b3b8ee29?collection=education},
-}
-
 @Article{Bessa2017,
   author   = {Bessa, Miguel and Bostanabad, R. and Liu, Z. and Hu, A. and Apley, Daniel W. and Brinson, C. and Chen, W. and Liu, Wing Kam},
   journal  = {Computer Methods in Applied Mechanics and Engineering},
@@ -34,31 +11,4 @@ @Article{Bessa2017
   doi      = {10.1016/j.cma.2017.03.037},
   file     = {:home/martin/Documents/Mendeley Desktop/Bessa et al. - 2017 - A framework for data-driven analysis of materials under uncertainty Countering the curse of dimensionality.pdf:pdf},
   keywords = {Design of experiments, Machine learning and data mining, Plasticity, Reduced order model, Self-consistent clustering analysis},
-}
-
-@Article{Bessa2019,
-  author   = {Bessa, Miguel and Glowacki, Piotr and Houlder, Michael},
-  journal  = {Advanced Materials},
-  title    = {{Bayesian Machine Learning in Metamaterial Design: Fragile Becomes Supercompressible}},
-  year     = {2019},
-  issn     = {15214095},
-  number   = {48},
-  pages    = {1--6},
-  volume   = {31},
-  abstract = {Designing future-proof materials goes beyond a quest for the best. The next generation of materials needs to be adaptive, multipurpose, and tunable. This is not possible by following the traditional experimentally guided trial-and-error process, as this limits the search for untapped regions of the solution space. Here, a computational data-driven approach is followed for exploring a new metamaterial concept and adapting it to different target properties, choice of base materials, length scales, and manufacturing processes. Guided by Bayesian machine learning, two designs are fabricated at different length scales that transform brittle polymers into lightweight, recoverable, and supercompressible metamaterials. The macroscale design is tuned for maximum compressibility, achieving strains beyond 94% and recoverable strengths around 0.1 kPa, while the microscale design reaches recoverable strengths beyond 100 kPa and strains around 80%. The data-driven code is available to facilitate future design and analysis of metamaterials and structures (https://github.com/mabessa/F3DAS).},
-  doi      = {10.1002/adma.201904845},
-  file     = {:home/martin/Documents/Mendeley Desktop/Bessa, Glowacki, Houlder - 2019 - Bayesian Machine Learning in Metamaterial Design Fragile Becomes Supercompressible.pdf:pdf},
-  keywords = {additive manufacturing, data-driven design, deep learning, machine learning, optimization},
-}
-
-@Article{Shin2022,
-  author    = {Shin, Dongil and Cupertino, Andrea and de Jong, Matthijs HJ and Steeneken, Peter G and Bessa, Miguel A and Norte, Richard A},
-  journal   = {Advanced Materials},
-  title     = {Spiderweb nanomechanical resonators via bayesian optimization: inspired by nature and guided by machine learning},
-  year      = {2022},
-  number    = {3},
-  pages     = {2106248},
-  volume    = {34},
-  eprint    = {2108.04809},
-  publisher = {Wiley Online Library},
-}
+}

paper/paper.md

@@ -7,10 +7,9 @@ tags:
   - framework
   - machine learning
 authors:
-  - name: Martin P. van der Schelling
+  - name: Martin van der Schelling
     orcid: 0000-0003-3602-0452
-    # equal-contrib: true
-    affiliation: 1 # (Multiple affiliations must be quoted)
+    affiliation: 1
   - name: Bernardo P. Ferreira
     orcid: 0000-0001-5956-3877
     affiliation: 2
@@ -29,23 +28,31 @@ bibliography: paper.bib
 ---
 
 # Summary
-<!-- A summary describing the high-level functionality and purpose of the software for a diverse, non-specialist audience. -->
 
 [`f3dasm`](https://github.com/bessagroup/f3dasm) (Framework for Data-driven Design and Analysis of Structures \& Materials) is a Python project that provides a general and user-friendly data-driven framework for researchers and practitioners working on the design and analysis of materials and structures. The package aims to streamline the data-driven process and make it easier to replicate research articles in this field, as well as share new work with the community. 
 
-![Logo of [`f3dasm`](https://github.com/bessagroup/f3dasm). \label{fig:f3dasm_logo}](f3dasm_logo.png)
+![Logo of [`f3dasm`](https://github.com/bessagroup/f3dasm). \label{fig:f3dasm_logo}](f3dasm_logo_long.png)
 
 # Statement of need
-<!-- A Statement of need section that clearly illustrates the research purpose of the software and places it in the context of related work. -->
 
 In the last decades, advancements in computational resources have accelerated novel inverse design approaches for structures and materials. In particular data-driven methods leveraging machine learning techniques play a major role in shaping our design processes today.
 
 Constructing a large material response database poses practical challenges, such as proper data management, efficient parallel computing and integration with third-party software. Because most applied fields remain conservative when it comes to openly sharing databases and software, a lot of research time is instead being allocated to implement common procedures that would be otherwise readily available. This lack of shared practices also leads to compatibility issues for benchmarking and replication of results by violating the FAIR principles.
 
-In this work we introduce an interface for researchers and practitioners working on design and analysis of materials and structures. The package is called [`f3dasm`](https://github.com/bessagroup/f3dasm) (Framework for Data-driven Design \& Analysis of Structures and Materials) This work generalizes the original closed-source framework proposed by the Bessa and co-workers [@Bessa2017], making it more flexible and adaptable to different applications, namely by allowing the integration of different choices of software packages needed in the different steps of the data-driven process: (1) design of experiments; (2) data generation; (3) machine learning; and (4) optimization. \autoref{fig:data-driven-process} provides an illustration of the stages in the data-driven process. 
+In this work we introduce an interface for researchers and practitioners working on design and analysis of materials and structures. The package is called [`f3dasm`](https://github.com/bessagroup/f3dasm) (Framework for Data-driven Design \& Analysis of Structures and Materials) This work generalizes the original closed-source framework proposed by the Bessa and co-workers [@Bessa2017], making it more flexible and adaptable to different applications, namely by allowing the integration of different choices of software packages needed in the different steps of the data-driven process:
+
+- **Design of experiments**, in which input variables describing the microstructure, properties and external conditions of the system are determined and sampled.
+- **Data generation**, typically through computational analyses, resulting in the creation of a material response database.
+- **Machine learning**, in which a surrogate model is trained to fit experimental findings.
+- **Optimization**, where we try to iteratively improve the design
+
+\autoref{fig:data-driven-process} provides an illustration of the stages in the data-driven process. 
 
 ![Illustration of the data-driven process. \label{fig:data-driven-process}](data-driven-process.png)
 
+
+[`f3dasm`](https://github.com/bessagroup/f3dasm) is an [open-source Python package](https://pypi.org/project/f3dasm/) compatible with Python 3.8 or later. The library includes a suite of benchmark functions, optimization algorithms, and sampling strategies to serve as default implementations. Furthermore, [`f3dasm`](https://github.com/bessagroup/f3dasm) offers automatic data management for experiments, easy integration with high-performance computing systems, and compatibility with the hydra configuration manager. Comprehensive [online documentation](https://f3dasm.readthedocs.io/en/latest/) is also available to assist users and developers of the framework.
+
 # Acknowledgements
 
 We would express our gratitude to Jiaxiang Yi for his contributions to writing an interface with the ABAQUS simulation software.