Pascal Pernot, Bing Huang, Andreas Savin B
Machine Learning Science and Technology·
2020·
cited by 26
Science & Research
Machine Learning in Materials Science
partially solved
importance 3/10
scale stated
Benchmarking quantum machine learning models with Mean Absolute Error learning curves can misrepresent true performance when error distributions are non-normal.
Michel Nass, Emil Alégroth, ROBERT H. FELDT
Software Testing Verification and Reliability·
2024·
cited by 16
Software & Coding
Software Testing and Debugging Techniques
ai assisted
importance 3/10
scale stated
Web-based test automation requires accurately locating web elements, but traditional attribute-matching methods miss context and meaning of elements and words.
Jiawei Li, David Faragó, Christian Petrov, Iftekhar Ahmed
Proceedings of the ACM on software engineering·
2024·
cited by 21
Software & Coding
Software Engineering Research
ai assisted
importance 3/10
scale stated
Automatically generating high-quality, informative commit messages for code changes, where prior approaches produced insufficient, poorly readable output.
Somayeh B. Shafiei, Saeed Shadpour, James Lloyd Mohler, Farzan Sasangohar, Camille Gutierrez, Mehdi Seilanian Toussi, Ambreen Shafqat
Journal of Robotic Surgery·
2023·
cited by 25
Medicine & Health
Surgical Simulation and Training
ai assisted
importance 3/10
scale stated
Surgical expertise in robot-assisted surgery is difficult to assess objectively; the study aimed to automatically classify surgeons' skill levels.
Andries Smit, Herman A. Engelbrecht, Willie Brink, Arnu Pretorius
Autonomous Agents and Multi-Agent Systems·
2023·
cited by 13
Robotics & Autonomy
Reinforcement Learning in Robotics
partially solved
importance 3/10
scale stated
Simulated RoboCup football teams still rely on handcrafted strategies, and reinforcement learning had not been scaled to the full 22-player game.
Fernando Díaz Ledezma, Sami Haddadin
Science Robotics·
2023·
cited by 18
Robotics & Autonomy
Robotic Mechanisms and Dynamics
partially solved
importance 3/10
scale stated
Robot morphology is typically assumed known and calibrated via external measuring devices; whether a robot can learn elements of its own body morphology by itself was largely unexplored.
Tjaša Rijavec, David Ribar, Jernej Markelj, Matija Strlič, Irena Kralj Cigić
Scientific Reports·
2022·
cited by 35
Identifying and quantifying plasticizers in historical PVC museum objects non-destructively was not previously done on a large, diverse collection.
Masaki Morimoto, Kai Fukami, Koji Fukagata
Physics of Fluids·
2021·
cited by 96
Physics & Astronomy
Model Reduction and Neural Networks
partially solved
importance 3/10
scale stated
Experimental particle image velocimetry often has missing regions in images due to experimental limitations, leaving gaps in measured velocity fields.
Emanuele Pesce, Giovanni Montana
Machine Learning·
2020·
cited by 49
Software & Coding
Reinforcement Learning in Robotics
partially solved
importance 3/10
scale stated
Multiple agents trained centrally but executed decentrally struggle to coordinate on tasks requiring synchronisation when they only have partial observations.
Nikita Moshkov, Botond Mathe, Attila Kertész‐Farkas, Réka Hollandi, Péter Horváth
Scientific Reports·
2020·
cited by 228
Deep learning-based cell segmentation on microscopy images requires massive annotated training sets to reach high accuracy; the paper addresses improving accuracy without more annotated data.
Wu Pang, Wang Jiangbei, Yanqiong Fei
Soft Robotics·
2017·
cited by 65
Steering an inchworm-inspired pneumatic soft robot is difficult because its motion is nonlinear, making closed-loop automatic control hard to achieve.
Darío G. Lema, Lídia Sánchez-González, Rubén Usamentiaga, Francisco J. delaCalle
Journal of Intelligent Manufacturing·
2025·
cited by 13
Comparing deep learning defect-detection models is unreliable because varying dataset partitions and stochastic training introduce variable outcomes.
Benjamin Luke Moorhouse, Tsz Ying Ho, Chenze Wu, Yuwei Wan
RELC Journal·
2025·
cited by 25
How pre-service language teachers can use large language models to improve lesson plans, and what knowledge and skills their task-specific prompting requires.
Γεώργιος Παπουτσόγλου, Sonia Tarazona, Marta Belchior Lopes, Thomas Klammsteiner, Eliana Ibrahimi, Julia Eckenberger, Pierfrancesco Novielli, Alberto Paolo Tonda et al.
Frontiers in Microbiology·
2023·
cited by 137
Microbiome ML workflows suffer domain-specific pitfalls in preprocessing, feature selection, modeling, and interpretation, risking unreliable disease diagnosis and biomarker discovery.
Sang-Hyun Park, Kang-Hee Lee, Ji Su Park, Youn-Soon Shin
Sustainability·
2022·
cited by 29
Engineering & Materials
Industrial Vision Systems and Defect Detection
partially solved
importance 3/10
scale stated
Product defects arise at several stages of manufacturing, and small-scale producers rely mainly on manual visual inspection to catch them.
Yang Zhang, Qingyu Yang, Dou An, Chengwei Zhang
Proceedings of the AAAI Conference on Artificial Intelligence·
2021·
cited by 20
Software & Coding
Reinforcement Learning in Robotics
partially solved
importance 3/10
scale stated
Existing multi-agent reinforcement learning methods only handle global- or neighborhood-level coordination, leaving coordination between individual agents unsolved in unknown multi-agent environments.
Daniel J. Cruz, Manuel R. Barbosa, Abel Dias dos Santos, Sara S. Miranda, Rui Amaral
Metals·
2021·
cited by 35
Manufacturers need to identify material constitutive parameters from bending tests and predict punch displacement to achieve a desired bending angle despite springback in press-brake air bending.
M. K. Cavanagh, K. Bekki
Astronomy and Astrophysics·
2020·
cited by 29
Physics & Astronomy
Galaxies: Formation, Evolution, Phenomena
ai assisted
importance 3/10
scale stated
Astronomers lacked understanding of how stellar bars form in galaxy merging and whether barred galaxies can be classified by their formation mechanism.
Isabelle Roesch, Tobias Günther
Computer Graphics Forum·
2018·
cited by 28
A single loss value hides variations in recurrent neural network weather forecasts, making it hard to tune hyperparameters and understand model behavior on large datasets.
Trisha A. Hinners, Kevin Tat, Rachel Thorp
The Astronomical Journal·
2018·
cited by 55
Physics & Astronomy
Stellar, planetary, and galactic studies
partially solved
importance 3/10
scale stated
Predicting and classifying stellar properties from noisy, sparse real Kepler light-curve time-series data, a task previously studied mostly on simulated data.
Alireza Bahramian
Surface and Interface Analysis·
2013·
cited by 20
Predicting the thickness of TiO2 nanostructured thin films based on sol-gel dip-coating process parameters.
Zeinab Shahbazi, Yung-Cheol Byun
Sensors·
2022·
cited by 34
Business & Industry
Stock Market Forecasting Methods
partially solved
importance 3/10
scale stated
Predicting cryptocurrency exchange rates is difficult because financial markets are non-linear and disordered.
Songjian Dan
International Journal on Semantic Web and Information Systems·
2022·
cited by 15
Identifying the geographic origin of oranges from NIR spectral data is needed for food authentication and quality control.
Guo Dong Goh, Xi Huang, Sheng Huang, Jia Li Janessa Thong, Jia Jun Seah, Wai Yee Yeong
Materials Science in Additive Manufacturing·
2023·
cited by 22
Additive manufacturing datasets for SLM Ti6Al4V are incomplete because individual studies cover only some parameters, hindering process-property linkage.