The potential of large language models for improving probability learning: A study on ChatGPT3.5 and first-year computer engineering students
Authors:
Angel Udias,
Antonio Alonso-Ayuso,
Ignacio Sanchez,
Sonia Hernandez,
Maria Eugenia Castellanos,
Raquel Montes Diez,
Emilio Lopez Cano
Abstract:
In this paper, we assess the efficacy of ChatGPT (version Feb 2023), a large-scale language model, in solving probability problems typically presented in introductory computer engineering exams. Our study comprised a set of 23 probability exercises administered to students at Rey Juan Carlos University (URJC) in Madrid. The responses produced by ChatGPT were evaluated by a group of five statistics…
▽ More
In this paper, we assess the efficacy of ChatGPT (version Feb 2023), a large-scale language model, in solving probability problems typically presented in introductory computer engineering exams. Our study comprised a set of 23 probability exercises administered to students at Rey Juan Carlos University (URJC) in Madrid. The responses produced by ChatGPT were evaluated by a group of five statistics professors, who assessed them qualitatively and assigned grades based on the same criteria used for students. Our results indicate that ChatGPT surpasses the average student in terms of phrasing, organization, and logical reasoning. The model's performance remained consistent for both the Spanish and English versions of the exercises. However, ChatGPT encountered difficulties in executing basic numerical operations. Our experiments demonstrate that requesting ChatGPT to provide the solution in the form of an R script proved to be an effective approach for overcoming these limitations. In summary, our results indicate that ChatGPT surpasses the average student in solving probability problems commonly presented in introductory computer engineering exams. Nonetheless, the model exhibits limitations in reasoning around certain probability concepts. The model's ability to deliver high-quality explanations and illustrate solutions in any programming language, coupled with its performance in solving probability exercises, suggests that large language models have the potential to serve as learning assistants.
△ Less
Submitted 9 October, 2023;
originally announced October 2023.
Properties of Nucleon Resonances by means of a Genetic Algorithm
Authors:
C. Fernandez-Ramirez,
E. Moya de Guerra,
A. Udias,
J. M. Udias
Abstract:
We present an optimization scheme that employs a Genetic Algorithm (GA) to determine the properties of low-lying nucleon excitations within a realistic photo-pion production model based upon an effective Lagrangian. We show that with this modern optimization technique it is possible to reliably assess the parameters of the resonances and the associated error bars as well as to identify weaknesse…
▽ More
We present an optimization scheme that employs a Genetic Algorithm (GA) to determine the properties of low-lying nucleon excitations within a realistic photo-pion production model based upon an effective Lagrangian. We show that with this modern optimization technique it is possible to reliably assess the parameters of the resonances and the associated error bars as well as to identify weaknesses in the models. To illustrate the problems the optimization process may encounter, we provide results obtained for the nucleon resonances $Δ$(1230) and $Δ$(1700). The former can be easily isolated and thus has been studied in depth, while the latter is not as well known experimentally.
△ Less
Submitted 24 June, 2008; v1 submitted 27 May, 2008;
originally announced May 2008.
