We study the foundations of variational inference, which frames posterior inference as an optimisation problem, for probabilistic programming. The dominant approach for optimisation in practice is stochastic gradient ...
详细信息
ISBN:
(数字)9783031300448
ISBN:
(纸本)9783031300431;9783031300448
We study the foundations of variational inference, which frames posterior inference as an optimisation problem, for probabilistic programming. The dominant approach for optimisation in practice is stochastic gradient descent. In particular, a variant using the so-called reparameterisation gradient estimator exhibits fast convergence in a traditional statistics setting. Unfortunately, discontinuities, which are readily expressible in programming languages, can compromise the correctness of this approach. We consider a simple (higher-order, probabilistic) programming language with conditionals, and we endow our language with both a measurable and a smoothed (approximate) value semantics. We present type systems which establish technical pre-conditions. Thus we can prove stochastic gradient descent with the reparameterisation gradient estimator to be correct when applied to the smoothed problem. Besides, we can solve the original problem up to any error tolerance by choosing an accuracy coefficient suitably. Empirically we demonstrate that our approach has a similar convergence as a key competitor, but is simpler, faster, and attains orders of magnitude reduction in work-normalised variance.
Genetic programming is an evolutionary approach known for its performance in program synthesis. However, it is not yet mature enough for a practical use in real-world software development, since usually many training ...
详细信息
ISBN:
(数字)9783031295737
ISBN:
(纸本)9783031295720;9783031295737
Genetic programming is an evolutionary approach known for its performance in program synthesis. However, it is not yet mature enough for a practical use in real-world software development, since usually many training cases are required to generate programs that generalize to unseen test cases. As in practice, the training cases have to be expensively hand-labeled by the user, we need an approach to check the program behavior with a lower number of training cases. Metamorphic testing needs no labeled input/output examples. Instead, the program is executed multiple times, first on a given (randomly generated) input, followed by related inputs to check whether certain user-defined relations between the observed outputs hold. In this work, we suggest MTGP, which combines metamorphic testing and genetic programming and study its performance and the generalizability of the generated programs. Further, we analyze how the generalizability depends on the number of given labeled training cases. We find that using metamorphic testing combined with labeled training cases leads to a higher generalization rate than the use of labeled training cases alone in almost all studied configurations. Consequently, we recommend researchers to use metamorphic testing in their systems if the labeling of the training data is expensive.
We researches a new iterative algorithm for quadratic programming problem based on the modified Uzawa algorithm and the ALPMHSS iteration algorithm in this paper. And we obtains the convergence theorem and the eigenva...
详细信息
The growing enrollments in computerscience courses and increase in class sizes necessitate scalable, automated tutoring solutions to adequately support student learning. While Large Language Models (LLMs) like GPT-4 ...
详细信息
In a recent study, students were periodically prompted to self-report engagement whileworking on computerprogramming assignments in a CS1 course. A regression model predicting time-on-task was proposed. While it was ...
详细信息
ISBN:
(纸本)9781450394314
In a recent study, students were periodically prompted to self-report engagement whileworking on computerprogramming assignments in a CS1 course. A regression model predicting time-on-task was proposed. While it was a significant improvement over ad-hoc estimation techniques, the study nevertheless suffered from lack of error analysis, lack of comparison with existing methods, subtle complications in prompting students, and small sample size. In this paper we report results from a study with an increased number of student participants and modified prompting scheme intended to better capture natural student behavior. Furthermore, we perform a cross-validation analysis on our refined regression model and present the resulting error bounds. We compare with threshold approaches and find that, in at least one context, a simple 5-minute threshold of inactivity is a reasonable estimate for whether a student is on-task or not. We show that our approach to modeling student engagement while programming is robust and suitable for identification of students in need of intervention, understanding engagement behavior, and estimating time taken on programming assignments.
Task planning for autonomous agents has typically been done using deep learning models and simulation- based reinforcement learning. This research proposes combining inductive learning techniques with goal-directed an...
详细信息
Task planning for autonomous agents has typically been done using deep learning models and simulation- based reinforcement learning. This research proposes combining inductive learning techniques with goal-directed answer set programming to increase the explainability and reliability of systems for task breakdown and completion. Preliminary research has led to the creation of a Python harness that utilizes s(CASP) to solve task problems in a computationally efficient way. Although this research is in the early stages, we are exploring solutions to complex problems in simulated task completion.
Constrained clustering problems have been studied extensively in recent years. In this paper, we focus on a class of constrained k-median problems with general constraints on facilities, denoted as GCF k-CMedian probl...
详细信息
Research into automated grading has increased as computerscience courses grow. Dynamic and static approaches are typically used to implement these graders, the most common implementation being unit testing to grade c...
详细信息
ISBN:
(纸本)9798400701382
Research into automated grading has increased as computerscience courses grow. Dynamic and static approaches are typically used to implement these graders, the most common implementation being unit testing to grade correctness. This paper expands upon an ongoing systematic literature review to provide an in-depth analysis of how machine learning (ML) has been used to grade and give feedback on programming assignments. We conducted a backward snowball search using the ML papers from an ongoing systematic review and selected 27 papers that met our inclusion criteria. After selecting our papers, we analysed the skills graded, the preprocessing steps, the ML implementation, and the models' evaluations. We find that most the models are implemented using neural network-based approaches, with most implementing some form of recurrent neural network (RNN), including Long Short-Term Memory, and encoder/decoder with attention mechanisms. Some graders implement traditional ML approaches, typically focused on clustering. Most ML-based automated grading, not many use ML to evaluate maintainability, readability, and documentation, but focus on grading correctness, a problem that dynamic and static analysis techniques, such as unit testing, rule-based program repair, and comparison to models or approved solutions, have mostly resolved. However, some ML-based tools, including those for assessing graphical output, have evaluated the correctness of assignments that conventional implementations cannot.
This paper is meant to be a survey about implicit characterizations of complexity classes by fragments of higher-order programming languages, with a special focus on type systems and subsystems of linear logic. Partic...
详细信息
This paper is meant to be a survey about implicit characterizations of complexity classes by fragments of higher-order programming languages, with a special focus on type systems and subsystems of linear logic. Particular emphasis will be put on Martin Hofmann's contributions to the subject, which very much helped in shaping the field.
In recent years, there has been evidence of a growing interest on the part of universities to know in advance the academic performance of their students and allow them to establish timely strategies to avoid desertion...
详细信息
In recent years, there has been evidence of a growing interest on the part of universities to know in advance the academic performance of their students and allow them to establish timely strategies to avoid desertion and failure. One of the biggest challenges to predicting student performance is presented in the course "programming Fundamentals" of computerscience, Software Engineering, and Information Systems Engineering careers in Peruvian universities for high student dropout rates. The objective of this research was to explore the efficiency of Long-Short Term Memory Networks (LSTM) in the field of Educational Data Mining (EDM) to predict the academic performance of students during the seventh, eighth, twelfth, and sixteenth weeks of the academic semester, which allowed us to identify students at risk of failing the course. This research compares several predictive models, such as Deep Neural Network (DNN), Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), Support Vector Classifier (SVM), and K-Nearest Neighbor (KNN). A major challenge machine learning algorithms face is a class imbalance in a dataset, resulting in over-fitting to the available data and, consequently, low accuracy. We use Generative Adversarial Networks (GAN) and Synthetic Minority Over-sampling Technique (SMOTE) to balance the data needed in our proposal. From the experimental results based on accuracy, precision, recall, and F1-Score, the superiority of our model is verified concerning a better classification, with 98.3% accuracy in week 8 using LSTM-GAN, followed by DNN-GAN with 98.1% accuracy.
暂无评论