This Work-in-Progress paper presents WebTA, a code critiquer developed using Java, currently being expanded to MATLAB for first-year engineering students. It outlines the modification process of the existing WebTA software for MATLAB and shares the results of three beta tests conducted with 52 students. The paper highlights the significance of providing an automated tool that helps students identify and improve coding errors is highlighted (addressing the gap between computer science and engineering education). WebTA's feedback-on-demand feature addresses the challenge of providing timely and personalized feedback to a large number of students, contributing to the enhancement of first-year engineering education.

This research is part of a larger development project that is working on a multi-programming language code critiquer called WebTA. The WebTA code-critiquing software is designed to be used in courses for novice programmers, e.g., CS1 a first engineering course. The authors report on a component of the project that makes initial steps towards a automating the identification of common student mistakes, or antipatterns in code. Antipatterns can be errors, inefficiencies, or incorrect style choices in the code. This works is aimed at Python and uses the machine learning algorithm, Random Forests, to identify a stylistic antipattern of crowded operators.

Several rule-based code critiquing systems have been developed to support programmers. However, these systems often are targeted toward experienced learners. Novice learners often lack self-efficacy in programming [1]-code critiquers targeted at novice programmers to promote student learning and enrich the overall education system are essential. Students' self-efficacy for programming is the perception of students' competence concerning programming [2], an essential attribute of computer science education. This paper examines existing literature on the impact of self-efficacy on students in programming. This work focuses on empirical work in programming education that independently addresses and develops theories specific to student programming-at the same time, addressing the existing gap in understanding the impact of code critiquers on student self-efficacy. The systematic review followed guidelines proposed by Kitchenham methodology. Findings revealed various factors that improve self-efficacy and provide evidence of significant sources of self-efficacy in programming. Moreover, the investigation guides further research in designing code critiquers to enhance the self-efficacy of novice learners. This work is part of a more significant effort to investigate the use of antipatterns (common programming mistakes) in novice programmers' coding. This sub-project aims to determine if the use of a code critiquer by first-year engineering students will improve student self-efficacy regarding programming.

Expanding autocritiquers to other languages requires the difficult task of obtaining and representing antipat-terns. Under the current framework, each language requires a unique Abstract Syntax Tree (AST) and error solution such that both the resulting AST and error summaries can be searched and presented to students. Often similar antipatterns arise that have to be rewritten in each language's own context. This paper proposes the creation of a generalized AST structure for an auto-critiquer under development, that can unify common code structures across similar languages while capturing differences in the languages. This work also proposes a standard for error message representation that can capture similar errors across language specific error messages. There are two motivations to standardizing AST and error message formats. Firstly, our solution allows for a standardized User Interface (UI) and easy integration of new languages with an identical on-boarding process for both professors and students. Secondly, reuse of antipatterns will allow us to skip the time-consuming step of constructing near-identical structural and logical antipattern queries as well as unifying novice based error feedback.

Modern architectures provide hardware memory prefetching capabilities which can be configured at runtime. While hardware prefetching can provide substantial performance improvements for many programs, prefetching can also increase contention for shared resources such as last-level cache and memory bandwidth. In turn, this contention can degrade performance in multi-core workloads. In this paper, we model fine-grained hardware prefetcher control as a contextual bandit, and propose a framework for learning prefetcher control policies which adjust hardware prefetching usage at runtime according to workload performance behavior. We train our policies on profiling data, wherein hardware memory prefetchers are enabled or disabled randomly at regular intervals over the course of a workload's execution. The learned prefetcher control policies provide up to a 4.3% average performance improvement over a set of memory bandwidth intensive workloads.

Virtualization technology is a key component for data center management which allows for multiple users and applications to share a single, physical machine. Modern virtual machine monitors utilize both software and hardware-assisted paging for memory virtualization, however neither paging mode is always preferable. Previous studies have shown that dynamic selection, which at runtime selects paging modes according to relevant performance metrics, can be effective in tailoring memory virtualization to program workload. However, these approaches require low-level manual analysis, or depend on prior knowledge of workload characteristics and phasing.

Students in computing disciplines need a strong basis in the fundamentals of discrete mathematics, but traditional offline approaches to teaching this material provide limited opportunities for the kind of interactive learning that computing students experience in their programming assignments. We have been using the Alloy language and analyzer to teach concepts in discrete structures (relational algebra, logic, and graphs) in an exploratory, programming-oriented way. Alloy, however, constitutes a new programming paradigm for introductory students, and careful mediation is needed to keep students on track. We use the familiar programming lab format, where students work on small-scope problems co-located with instructors, to provide guidance as students wrestle with the languages of relational algebra and predicate logic through Alloy. We describe selected lab exercises, and report on initial findings based on our experiences with students.

The field of AI has changed significantly in the past couple of years and will likely continue to do so. Driven by a desire to expose our students to relevant and modern materials, we conducted two surveys, one of AI instructors and one of AI practitioners. The surveys were aimed at gathering infor-mation about the current state of the art of introducing AI as well as gathering input from practitioners in the field on techniques used in practice. In this paper, we present and briefly discuss the responses to those two surveys.

Today’s data centers are designed with multi-core CPUs where multiple virtual machines (VMs) can be co-located into one physical machine or distribute multiple computing tasks onto one physical machine. The result is co-tenancy, resource sharing and competition. Modeling and predicting such co-run interference becomes crucial for job scheduling and Quality of Service assurance. Co-locating interference can be characterized into two components, sensitivity and pressure, where sensitivity characterizes how an application’s own performance is affected by a co-run application, and pressure characterizes how much contentiousness an application exerts/brings onto the memory subsystem. Previous studies show that with simple models, sensitivity and pressure can be accurately characterized for a single machine. We extend the models to consider cross-architecture sensitivity (across different machines).

Cloud computing has become a dominant computing paradigm to provide elastic, affordable computing resources to end users. Due to the increased computing power of modern machines powered by multi/many-core computing, data centers often co-locate multiple virtual machines (VMs) into one physical machine, resulting in co-tenancy, and resource sharing and competition. Applications or VMs co-locating in one physical machine can interfere with each other despite of the promise of performance isolation through virtualization. Modelling and predicting co-run interference therefore becomes critical for data center job scheduling and QoS (Quality of Service) assurance. Co-run interference can be categorized into two metrics, sensitivity and pressure, where the former denotes how an application's performance is affected by its co-run applications, and the latter measures how it impacts the performance of its co-run applications. This paper shows that sensitivity and pressure are both application-and architecture dependent. Further, we propose a regression model that predicts an application's sensitivity and pressure across architectures with high accuracy. This regression model enables a data center scheduler to guarantee the QoS of a VM/application when it is scheduled to co-locate with another VMs/applications.

The Model AI Assignments session seeks to gather and disseminate the best assignment designs of the Artificial Intelligence (AI) Education community. Recognizing that assignments form the core of student learning experience, we here present abstracts of five AI assignments from the 2014 session that are easily adoptable, playfully engaging, and flexible for a variety of instructor needs. Assignment specifications and supporting resources may be found at http://modelai.gettysburg.edu.

The Symposium on Educational Advances in Artificial Intelligence (EAAI-13) seeks to advance the AAAI goal of improving the teaching and training of AI practitioners. The symposium provides a venue for both researchers and educators to discuss pedagogical issues and share resources related to teaching AI and using AI in education across a variety of curricular levels (K-12 through postgraduate training).

In the 2012-2013 academic year, a survey to investigate why engineering and computer science students persist in their major was conducted at Michigan Technological University. This paper discusses the results of the survey and ties the findings to the literature. It focuses on: (1) who influenced students' decisions on picking a major or on changing a major (for example, friends, family, academic advisors, faculty, upper-division or graduate students, co-workers, and supervisors), and how did they affect students' persistence and (2) what is the impact of role models on student persistence. The analysis compares students who reported not having considered changing majors to students who considered switching to another major. The findings show that the students who did not consider changing majors reported having a stronger support system including faculty, academic advisors, and engineers who serve as role models. The data suggest that university faculty and staff need to reach out to the students who are deliberating about their initial choice of major and support the decision making process.

Polynomial Support Vector Machine models of degree d are linear functions in a feature space of monomials of at most degree d. However, the actual representation is stored in the form of support vectors and Lagrange multipliers that is unsuitable for

We present a new algorithm for Bayesian network structure learning, called Max-Min Hill-Climbing (MMHC). The algorithm combines ideas from local learning, constraint-based, and search-and-score techniques in a principled and effective way. It first reconstructs the skeleton of a Bayesian network and then performs a Bayesian-scoring greedy hill-climbing search to orient the edges. In our extensive empirical evaluation MMHC outperforms on average and in terms of various metrics several prototypical and state-of-the-art algorithms, namely the PC, Sparse Candidate, Three Phase Dependency Analysis, Optimal Reinsertion, Greedy Equivalence Search, and Greedy Search. These are the first empirical results simultaneously comparing most of the major Bayesian network algorithms against each other. MMHC offers certain theoretical advantages, specifically over the Sparse Candidate algorithm, corroborated by our experiments. MMHC and detailed results of our study are publicly available at http://www.dsl-lab.org/supplements/mmhc_paper/mmhc_index.html.