Search

In the ever-evolving landscape of machine learning, seamless translation of natural language descriptions into executable code remains a formidable challenge. This article introduces Linguacodus, an innovative framework designed to tackle this challenge by deploying a dynamic pipeline that iteratively transforms natural language task descriptions into code through high-level data-shaping instructions. The core of Linguacodus is a fine-tuned large language model, empowered to evaluate diverse solutions for various problems and select the most fitting one for a given task. This article details the fine-tuning process and sheds light on how natural language descriptions can be translated into functional code. Linguacodus represents a substantial leap towards automated code generation, effectively bridging the gap between task descriptions and executable code. It holds great promise for advancing machine learning applications across diverse domains. Additionally, we propose an algorithm capable of transforming a natural description of an ML task into code with minimal human interaction. In extensive experiments on a vast machine learning code dataset originating from Kaggle, we showcase the effectiveness of Linguacodus. The investigations highlight its potential applications across diverse domains, emphasizing its impact on applied machine learning in various scientific fields. [ABSTRACT FROM AUTHOR]

The article introduces a model-based framework, FAIR-Frame, for assessing bias in machine learning pipelines, emphasizing the need to consider both upstream representational harm and downstream allocational harm. Topics include the limitations of existing fairness metrics, the proposed framework's evaluation on text classification tasks, and its implications for various machine learning contexts.

In the ever-evolving landscape of machine learning, seamless translation of natural language descriptions into executable code remains a formidable challenge. This article introduces Linguacodus, an innovative framework designed to tackle this challenge by deploying a dynamic pipeline that iteratively transforms natural language task descriptions into code through high-level data-shaping instructions. The core of Linguacodus is a fine-tuned large language model, empowered to evaluate diverse solutions for various problems and select the most fitting one for a given task. This article details the fine-tuning process and sheds light on how natural language descriptions can be translated into functional code. Linguacodus represents a substantial leap towards automated code generation, effectively bridging the gap between task descriptions and executable code. It holds great promise for advancing machine learning applications across diverse domains. Additionally, we propose an algorithm capable of transforming a natural description of an ML task into code with minimal human interaction. In extensive experiments on a vast machine learning code dataset originating from Kaggle, we showcase the effectiveness of Linguacodus. The investigations highlight its potential applications across diverse domains, emphasizing its impact on applied machine learning in various scientific fields.

Domain experts are increasingly employing machine learning to solve their domain-specific problems. This article presents to software engineering researchers the six key challenges that a domain expert faces in addressing their problem with a computational workflow, and the underlying executable implementation. These challenges arise out of our conceptual framework which presents the "route" of transformations that a domain expert may choose to take while developing their solution. To ground our conceptual framework in the state of the practice, this article discusses a selection of available textual and graphical workflow systems and their support for the transformations described in our framework. Example studies from the literature in various domains are also examined to highlight the tools used by the domain experts as well as a classification of the domain specificity and machine learning usage of their problem, workflow, and implementation. The state of the practice informs our discussion of the six key challenges, where we identify which challenges and transformations are not sufficiently addressed by available tools. We also suggest possible research directions for software engineering researchers to increase the automation of these tools and disseminate best-practice techniques between software engineering and various scientific domains. [ABSTRACT FROM AUTHOR]

Long-term river streamflow prediction and modeling are essential for water resource management and decision-making related to water resources. This research paper considers the importance of these predictions and proposes a model to address scarcity scenarios to support decision-making in water allocation, flood management, and drought prediction scenarios. Machine learning (ML) techniques offer promising alternatives for improving long-term streamflow prediction. However, most existing studies on ML models for streamflow prediction have focused on shorter time horizons, limiting their broader applicability. Consequently, there is a need for dedicated research that addresses the use of ML models in long-term streamflow prediction. Considering this research gap, this paper presents an ML-based approach that learns and replicates the natural flow dynamics of a river, allowing for the simulation of reduced flow scenarios (25 % and 50 % reduction). This capability allows for simulating drought scenarios of varying severity, providing valuable insights for water service managers. This study significantly contributes to the progress of predicting long-term river streamflow through the application of machine learning models. Moreover, this study offers valuable insights and recommendations for hydrologists to improve future research efforts.

The automation of data science and other data manipulation processes depend on the integration and formatting of 'messy' data. Data wrangling is an umbrella term for these tedious and time-consuming tasks. Tasks such as transforming dates, units or names expressed in different formats have been challenging for machine learning because (1) users expect to solve them with short cues or few examples, and (2) the problems depend heavily on domain knowledge. Interestingly, large language models today (1) can infer from very few examples or even a short clue in natural language, and (2) can integrate vast amounts of domain knowledge. It is then an important research question to analyse whether language models are a promising approach for data wrangling, especially as their capabilities continue growing. In this paper we apply different variants of the language model Generative Pre-trained Transformer (GPT) to five batteries covering a wide range of data wrangling problems. We compare the effect of prompts and few-shot regimes on their results and how they compare with specialised data wrangling systems and other tools. Our major finding is that they appear as a powerful tool for a wide range of data wrangling tasks. We provide some guidelines about how they can be integrated into data processing pipelines, provided the users can take advantage of their flexibility and the diversity of tasks to be addressed. However, reliability is still an important issue to overcome. [ABSTRACT FROM AUTHOR]

Machine Learning allows facing complex tasks related to data analysis with big datasets. This Artificial Intelligence branch allows not technical contexts to get benefits related to data processing and analysis. In particular, in medicine, medical professionals are increasingly interested in Machine Learning to identify patterns in clinical cases and make predictions regarding health issues. However, many do not have the necessary programming or technological skills to perform these tasks. Many different tools focus on developing Machine Learning pipelines, from libraries for developers and data scientists to visual tools for experts or platforms to learn. However, we have identified some requirements in the medical context that raise the need to create a customized platform adapted to end-user found in this context. This work describes the design process and the first version of KoopaML, an ML platform to bridge the data science gaps of physicians while automatizing Machine Learning pipelines. The platform is focused on enhanced interactivity to improve the engagement of physicians while still providing all the benefits derived from the introduction of Machine Learning pipelines in medical departments, as well as integrated ongoing training during the use of the tool’s features.

Machine learning (ML) is increasingly used to automate impactful decisions, and the risks arising from this widespread use are garnering attention from policy makers, scientists, and the media. ML applications are often brittle with respect to their input data, which leads to concerns about their correctness, reliability, and fairness. In this paper, we describe mlinspect, a library that helps diagnose and mitigate technical bias that may arise during preprocessing steps in an ML pipeline. We refer to these problems collectively as data distribution bugs. The key idea is to extract a directed acyclic graph representation of the dataflow from a preprocessing pipeline and to use this representation to automatically instrument the code with predefined inspections. These inspections are based on a lightweight annotation propagation approach to propagate metadata such as lineage information from operator to operator. In contrast to existing work, mlinspect operates on declarative abstractions of popular data science libraries like estimator/transformer pipelines and does not require manual code instrumentation. We discuss the design and implementation of the mlinspect library and give a comprehensive end-to-end example that illustrates its functionality. [ABSTRACT FROM AUTHOR]

Recent years have seen a proliferation of ML frameworks. Such systems make ML accessible to non-experts, especially when combined with powerful parameter tuning and AutoML techniques. Modern, applied ML extends beyond direct learning on clean data, however, and needs an expressive language for the construction of complex ML workflows beyond simple pre- and post-processing. We present mlr3pipelines, an R framework which can be used to define linear and complex non-linear ML workflows as directed acyclic graphs. The framework is part of the mlr3 ecosystem, leveraging convenient resampling, benchmarking, and tuning components. [ABSTRACT FROM AUTHOR]

Machine Learning (ML) has gained prominence and has tremendous applications in fields like medicine, biology, geography and astrophysics, to name a few. Arguably, in such areas, it is used by domain experts, who are not necessarily skilled-programmers. Thus, it presents a steep learning curve for such domain experts in programming ML applications. To overcome this and foster widespread adoption of ML techniques, we propose to equip them with domain-specific graphical tools. Such tools, based on the principles of flow-based programming paradigm, would support the graphical composition of ML applications at a higher level of abstraction and auto-generation of target code. Accordingly, (i) we have modelled ML algorithms as composable components; (ii) described an approach to parse a flow created by connecting several such composable components and use an API-based code generation technique to generate the ML application. To demonstrate the feasibility of our conceptual approach, we have modelled the APIs of Apache Spark ML as composable components and validated it in three use-cases. The use-cases are designed to capture the ease of program specification at a higher abstraction level, easy parametrisation of ML APIs, auto-generation of the ML application and auto-validation of the generated model for better prediction accuracy.

Machine Learning (ML) has gained prominence and has tremendous applications in fields like medicine, biology, geography and astrophysics, to name a few. Arguably, in such areas, it is used by domain experts, who are not necessarily skilled-programmers. Thus, it presents a steep learning curve for such domain experts in programming ML applications. To overcome this and foster widespread adoption of ML techniques, we propose to equip them with domain-specific graphical tools. Such tools, based on the principles of flow-based programming paradigm, would support the graphical composition of ML applications at a higher level of abstraction and auto-generation of target code. Accordingly, (i) we have modelled ML algorithms as composable components; (ii) described an approach to parse a flow created by connecting several such composable components and use an API-based code generation technique to generate the ML application. To demonstrate the feasibility of our conceptual approach, we have modelled the APIs of Apache Spark ML as composable components and validated it in three use-cases. The use-cases are designed to capture the ease of program specification at a higher abstraction level, easy parametrisation of ML APIs, auto-generation of the ML application and auto-validation of the generated model for better prediction accuracy. [ABSTRACT FROM AUTHOR]