We study the convergence dynamics of Gradient Descent (GD) in a minimal binary classification setting, consisting of a two-neuron ReLU network and two training instances. We prove that even under these strong simplifying assumptions, while GD successfully converges to an optimal robustness margin, effectively maximizin...
Guy Smorodinsky, Sveta Gimpleson, Itay Safran· 0 citations
We investigate the theoretical aspects of offline reinforcement learning (RL) under general function approximation. While prior works (e.g., Xie et al., 2021) have established the theoretical foundations of learning a good policy from offline data via pessimism, existing algorithms that are computationally tractable (o...
Training large language models (LLMs) for multi-turn code generation typically relies on online reinforcement learning (RL), in which models learn from repeatedly generated online rollouts. Although effective, online RL can be costly and unstable. In this paper, we present contextual bandit learning with offline trajec...
Ziru Chen, Dongdong Chen, Ruinan Jin et al.· 0 citations
A central difficulty in applying reinforcement learning to real-world applications such as autonomous driving is that conventional scalar reward formulations obtained by aggregating multiple often competing objectives may obscure the structure of their trade-offs. These challenges are particularly relevant to decision-...
Deepthi Pathare, Leo Laine, Morteza Haghir Chehreghani· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Automated program repair (APR) has recently shifted toward large language models and agent-based systems, while prior work shows that repository history can help repair single-line bugs. We conduct a large-scale empirical study of whether, how much, and under what conditions repository history improves agentic APR beyo...
Traditional machine learning relies on explicit models and domain assumptions, often limiting flexibility and interpretability. We introduce a "model-free" framework of instance-based learning using surprisal (information theoretic uncertainty) to directly analyze and perform inferences from raw data, eliminating distr...
Christopher J. Hazard, Michael Resnick, Jacob Beel et al.· 0 citations
Large language models (LLMs) are increasingly under scrutiny for perpetuating identity-based discrimination in high-stakes domains such as hiring, particularly against people with disabilities (PwD). However, existing research remains largely Western-centric, overlooking how intersecting forms of marginalization--such...
Mahika Phutane, Hayoung Jung, Matthew Kim et al.· 0 citations
Mechanics-inspired neural dynamical models improve interpretability and physical consistency, but they usually assume a known configuration space, fixed physical parameters, and deterministic prediction from a prescribed initial condition. We introduce Structured Phase Space GAN (SPS-GAN), a conditional generative mode...
Yulong Yang, Jiayin Liu, Vineet Bansal et al.· 0 citations
A classic account of the quantification of semantic information is that of Bar-Hiller and Carnap. Their account proposes an inverse relation between the informativeness of a statement and its probability. However, their approach assigns the maximum informativeness to a contradiction: which Floridi refers to as the Bar-...
With origins in game theory, probabilistic values like Shapley values, Banzhaf values, and semi-values have emerged as a central tool in explainable AI. They are used for feature attribution, data attribution, data valuation, and more. Since exact computation can require exponential time, research has focused on effici...
R. Teal Witter, Yurong Liu, Christopher Musco· 0 citations
Indirect prompt injection attacks exploit a fundamental weakness of LLMs: executing unauthorized tasks from untrusted sources. Despite the practical significance of this attack vector, existing benchmarks rely on small sets of template-based attacks (100-1,000 examples) that fail to capture the diversity and adaptivene...
Sahar Abdelnabi, Aideen Fay, Ahmed Salem et al.· 0 citations
Generative models have become increasingly powerful tools for robot motion generation, enabling flexible and multimodal trajectory generation across various tasks. Yet, most existing approaches remain limited in handling multiple types of constraints, such as collision avoidance, actuation limits, and dynamic consisten...
Zewen Yang, Xiaobing Dai, Dian Yu et al.· 0 citations
