Related papers: A rapid evaluation of Australia's COVID-era appren…
Recently, contrastive learning has risen to be a promising approach for large-scale self-supervised learning. However, theoretical understanding of how it works is still unclear. In this paper, we propose a new guarantee on the downstream…
The COVID-19 pandemic has represented a challenge for higher education in terms to provide quality education despite the lockdown periods, the transformation of the in-person classes to virtual classes, and the demotivation and anxiety that…
Rewards and punishments in different forms are pervasive and present in a wide variety of decision-making scenarios. By observing the outcome of a sufficient number of repeated trials, one would gradually learn the value and usefulness of a…
We present results of an experiment benchmarking a workforce training program against cash transfers for underemployed young adults in Rwanda. 3.5 years after treatment, the training program enhances productive time use and asset…
Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs. We present BoostAPR, a three-stage framework addressing these challenges: (1)…
Open-ended programming increases students' motivation by allowing them to solve authentic problems and connect programming to their own interests. However, such open-ended projects are also challenging, as they often encourage students to…
With the growth of interest in the attack and defense of deep neural networks, researchers are focusing more on the robustness of applying them to devices with limited memory. Thus, unlike adversarial training, which only considers the…
Real-world control systems frequently operate under \emph{piecewise stationary} conditions, where dynamics remain stable for extended periods before undergoing abrupt regime changes. Standard robust RL methods face a fundamental dilemma: a…
We evaluate two interventions facilitating technology-sector transitions for women in Poland: Mentoring, focused on expanding professional networks, and Challenges, focused on building credible skill signals. Randomizing oversubscribed…
Experimental evidence confirms that AI tools raise worker productivity, but also that sustained use can erode the expertise on which those gains depend. We develop a dynamic model in which a decision-maker chooses AI usage intensity for a…
The Pay-as-Clear (PaC) mechanism currently used in the European electricity market can generate significant submarginal profits for renewable sources when the clearing price is determined by the marginal offers of gas-fired generation units…
Marketing and product personalisation provide a prominent and visible use-case for the application of Information Retrieval methods across several business domains. Recently, agentic approaches to these problems have been gaining traction.…
Training-time safety violations have been a major concern when we deploy reinforcement learning algorithms in the real world. This paper explores the possibility of safe RL algorithms with zero training-time safety violations in the…
Assessments help in evaluating the knowledge gained by a learner at any specific point as well as in continuous improvement of the curriculum design and the whole learning process. However, with the increase in students' enrollment at…
Given the increasing scale of model sizes, novel training strategies like gradual stacking [Gong et al., 2019, Reddi et al., 2023] have garnered interest. Stacking enables efficient training by gradually growing the depth of a model in…
The success of reinforcement learning (RL) is fundamentally tied to having a reward function that accurately reflects the task objective. Yet, designing reward functions is notoriously time-consuming and prone to misspecification. To…
The COVID-19 pandemic has permanently altered workplace structures, normalizing remote work. However, critical evidence highlights challenges with fully remote arrangements, particularly for software teams. This study investigates employee…
Dataset bias is one of the prevailing causes of unfairness in machine learning. Addressing fairness at the data collection and dataset preparation stages therefore becomes an essential part of training fairer algorithms. In particular,…
The increasing demand for programmers has led to a surge in participants in programming courses, making it increasingly challenging for instructors to assess student code manually. As a result, automated programming assessment systems…
Computer science (CS) capstone courses offer students a valuable opportunity to gain hands-on experience in software development, practice essential soft skills, and enhance their employability prospects. They are a core component in many…