Machine Learning Operations

Soh, Julian; Singh, Priyanshi

doi:10.1007/978-1-4842-6405-8_8

Cited by 13 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding the ML model, an initial version is trained with the whole dataset (i.e., input variables annotated with some clinical label) when the best pipeline and configuration (or “hyperparameters”) have been selected, before being deployed in real conditions and confronted to new data ( 54 ). Thus, the training of an initial model is undertaken systematically.…”

Section: Discussionmentioning

confidence: 99%

“…Regarding the ML model, when the pipeline is put into production (real-life situation), no background modification is possible (the pipeline is static—it has been chosen during the training phase), but a shift of specialization is possible depending on the data it is fed with ( 54 , 56 ).…”

Section: Discussionmentioning

confidence: 99%

“…Regarding ML models, when they are put into production, the only parameter that could be set regarding their degree of freedom is the choice to fine train the model with the new samples, and if so, the importance to give to them ( 54 , 56 ).…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

How Does Comparison With Artificial Intelligence Shed Light on the Way Clinicians Reason? A Cross-Talk Perspective

et al. 2022

View full text Add to dashboard Cite

In order to create a dynamic for the psychiatry of the future, bringing together digital technology and clinical practice, we propose in this paper a cross-teaching translational roadmap comparing clinical reasoning with computational reasoning. Based on the relevant literature on clinical ways of thinking, we differentiate the process of clinical judgment into four main stages: collection of variables, theoretical background, construction of the model, and use of the model. We detail, for each step, parallels between: i) clinical reasoning; ii) the ML engineer methodology to build a ML model; iii) and the ML model itself. Such analysis supports the understanding of the empirical practice of each of the disciplines (psychiatry and ML engineering). Thus, ML does not only bring methods to the clinician, but also supports educational issues for clinical practice. Psychiatry can rely on developments in ML reasoning to shed light on its own practice in a clever way. In return, this analysis highlights the importance of subjectivity of the ML engineers and their methodologies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

How Does Comparison With Artificial Intelligence Shed Light on the Way Clinicians Reason? A Cross-Talk Perspective

et al. 2022

View full text Add to dashboard Cite

show abstract

“…For the selection of soil samples to train the machine learning models, we cleaned and de-noized the data which may significantly affect the model's performance (Soh & Singh, 2020). This included removing points that are spatial outliers and inconsistent with the description of sampling sites and data derived from a LiDAR DEM.…”

Section: Automated Machine Learningmentioning

confidence: 99%

Effects of spatial data resolution on the modelling and mapping of soil organic carbon content in hill country grassland landscapes

Tran,

Dominati,

Lowry

et al. 2023

Soil Use and Management

View full text Add to dashboard Cite

Limited use has been made of spatially explicit modelling of soil organic carbon (SOC) in highly complex farmed landscapes to advance current mapping efforts. This study aimed to address this gap in knowledge by evaluating the spatial prediction of SOC content in the 0‐75 mm soil depth in hill country landscapes in New Zealand (NZ) using point‐based training data, along with topographic covariates and Sentinel 2 spectral band ratios using an automated set of machine learning (AutoML) tools in ArcGIS. Subsequently, it also focused on quantifying the effects of spatial data resolution (i.e., 1m, 8m, 15m, and 25m) in terms of predicted map accuracy. Farmlets with contrasting phosphorus fertiliser and sheep grazing histories located at the Ballantrae Hill Country Research Station, NZ were selected to conduct the research. Six candidate algorithms incorporated in the AutoML tools (i.e., XGBoost, LightGBM, linear regression, decision trees, extra trees, random forest) and ensemble model were utilised to model the spatial pattern of SOC content. The results show that the ensemble model that combine predictions of various algorithms applied for 1m data resolution enables the highest performance and accuracy (i.e., R2 = 0.76, RMSE = 0.66%). Among the predictive variables used in the model, slope, wetness, and topographic position indices were found to be the most important topographical features that explain SOC patterns in the study area. Inclusion of spectral indices derived from remote sensing, including surface soil moisture and clay minerals ratio, made further improvement to the SOC content prediction. The study reveals that a decrease in the resolution of the geospatial data does not substantively affect the mean SOC content estimation of a farm‐scale modelling. However, using coarser resolution data reduces the ability of the model to predict changes in the spatial pattern of SOC content across a hill country grassland landscape.

show abstract

“…This is similar to the way DevOps helps software engineers develop, test and deploy software faster and with fewer defects. MLOps supports the data science life cycle in the same way that DevOps supports the application development lifecycle [47].…”

Section: Mlopsmentioning

confidence: 99%

Design Methods and Processes for ML/DL models

John¹

View full text Add to dashboard Cite

Context: With the advent of Machine Learning (ML) and especially Deep Learning (DL) technology, companies are increasingly using Artificial Intelligence (AI) in systems, along with electronics and software. Nevertheless, the end-to-end process of developing, deploying and evolving ML and DL models in companies brings some challenges related to the design and scaling of these models. For example, access to and availability of data is often challenging, and activities such as collecting, cleaning, preprocessing, and storing data, as well as training, deploying and monitoring the model(s) are complex. Regardless of the level of expertise and/or access to data scientists, companies in all embedded systems domain struggle to build high-performing models due to a lack of established and systematic design methods and processes. Objective: The overall objective is to establish systematic and structured design methods and processes for the end-to-end process of developing, deploying and successfully evolving ML/DL models. Method: To achieve the objective, we conducted our research in close collaboration with companies in the embedded systems domain using different empirical research methods such as case study, action research and literature review. Results and Conclusions: This research provides six main results: First, it identifies the activities that companies undertake in parallel to develop, deploy and evolve ML/DL models, and the challenges associated with them. Second, it presents a conceptual framework for the continuous delivery of ML/DL models to accelerate AI-driven business in companies. Third, it presents a framework based on current literature to accelerate the end-to-end deployment process and advance knowledge on how to integrate, deploy and operationalize ML/DL models. Fourth, it develops a generic framework with five architectural alternatives for deploying ML/DL models at the edge. These architectural alternatives range from a centralized architecture that prioritizes (re)training in the cloud to a decentralized architecture that prioritizes (re)training at the edge. Fifth, it identifies key factors to help companies decide which architecture to choose for deploying ML/DL models. Finally, it explores how MLOps, as a practice that brings together data scientist teams and operations, ensures the continuous delivery and evolution of models.

show abstract

Machine Learning Operations

Cited by 13 publications

References 0 publications

How Does Comparison With Artificial Intelligence Shed Light on the Way Clinicians Reason? A Cross-Talk Perspective

How Does Comparison With Artificial Intelligence Shed Light on the Way Clinicians Reason? A Cross-Talk Perspective

Effects of spatial data resolution on the modelling and mapping of soil organic carbon content in hill country grassland landscapes

Design Methods and Processes for ML/DL models

Contact Info

Product

Resources

About