Enoch Hyunwook Kang

Enoch Hyunwook Kang

In the forward reinforcement-learning problem, the reward is fixed and known; the learner is asked to find a good policy or value function. Here we turn the question around. Given offline data generated by an expert, can we recover the reward the expert was optimizing? This is the inverse reinforcement learning problem, and remarkably, two communities, structural econometricians studying dynamic discrete choice (DDC) and machine learners studying entropy-regularized IRL, have been working on exactly the same probabilistic model under different names. We begin by proving their equivalence. We then develop the classical identification result of Magnac and Thesmar and the classical computational paradigms that grew out of it: Rust's nested fixed-point algorithm, the conditional-choice-probability approach of Hotz and Miller, and the two temporal-difference approaches of Adusumilli and Eckardt: linear semi-gradient TD and approximate value iteration. Each route has its limits: dimensionality, transition-kernel estimation, the deadly triad, or projected fixed-point bias. We then walk through the modern ML/IRL strand: adversarial IRL, occupancy matching, IQ-Learn, and offline ML-IRL, deriving each method's actual objective and stating precisely what it does and does not identify. We close with the empirical-risk-minimization framework of Kang et al., which yields a gradient-based estimator for offline IRL/DDC.

Enoch Hyunwook Kang, Hema Yoganarasimhan

Large Language Models (LLMs) have recently enabled self-improving AI, i.e., AI that iteratively generates, evaluates, and refines its own outcomes. Recent studies have shown that self-improving AI focusing on prompt optimization can outperform state-of-the-art reinforcement-learning fine-tuned LLMs. Here, their `performance' is typically measured by query efficiency - the number of LLM-generated solution samples required to meet a certain performance threshold. However, in many societal applications, the primary limitation is not generating new solutions but evaluating them. For instance, evaluating an ad's effectiveness requires significant human feedback, which is far more costly and time-consuming than generating a candidate ad. To optimize for the evaluation efficiency objective, a natural approach is to extend Bayesian Optimization (BO), a framework proven optimal for evaluation efficiency, to the language domain. However, the difficulty of directly estimating suitable acquisition functions in LLMs' minds makes this extension challenging. This paper overcomes this challenge by proving that the combination of the simple and widely used Best-of-N selection strategy and simple textual gradients (i.e., textual edits from a critic model) statistically emulates the behavior of the gradients on the canonical UCB acquisition function, which induces optimal exploration in terms of evaluation efficiency. Based on this result, we propose TextGrad-Best-of-N Bayesian Optimization (T-BoN BO), a simple and eval-efficient language-space Bayesian optimization framework for AI self-improvement. We also empirically validate T-BoN BO by applying it to automated ad alignment tasks for persona distribution, demonstrating its superior performance compared to popular state-of-the-art baselines.

Enoch Hyunwook Kang, P. R. Kumar

In online exploration systems where users with fixed preferences repeatedly arrive, it has recently been shown that O(1), i.e., bounded regret, can be achieved when the system is modeled as a linear contextual bandit. This result may be of interest for recommender systems, where the popularity of their items is often short-lived, as the exploration itself may be completed quickly before potential long-run non-stationarities come into play. However, in practice, exact knowledge of the linear model is difficult to justify. Furthermore, potential existence of unobservable covariates, uneven user arrival rates, interpretation of the necessary rank condition, and users opting out of private data tracking all need to be addressed for practical recommender system applications. In this work, we conduct a theoretical study to address all these issues while still achieving bounded regret. Aside from proof techniques, the key differentiating assumption we make here is the presence of effective Synthetic Control Methods (SCM), which are shown to be a practical relaxation of the exact linear model knowledge assumption. We verify our theoretical bounded regret result using a minimal simulation experiment.

Enoch Hyunwook Kang

Personalized alignment aims to adapt large language models to heterogeneous user preferences, yet the precise theoretical conditions for its statistical efficiency have not been formally established. This paper characterizes the conditions under which personalized alignment achieves O(1) online regret and log(1/epsilon) offline sample complexity. We show that these optimal rates depend on a specific user-diversity condition: the population of user-specific heads must span the latent reward directions that can alter the optimal response. We prove that this condition is both necessary and sufficient. When it holds, simple greedy algorithms achieve benchmark efficiency; when it fails, every learner in a natural admissible class incurs at least logarithmic regret. Our results identify user diversity as the fundamental driver of personalized identifiability.

Enoch Hyunwook Kang

Iterative alignment methods based on purely greedy updates are remarkably effective in practice, yet existing theoretical guarantees of \(O(\log T)\) KL-regularized regret can seem pessimistic relative to their empirical performance. In this paper, we argue that this mismatch arises from the regret criterion itself: KL-regularized regret conflates the statistical cost of learning with the exploratory randomization induced by the softened training policy. To separate these effects, we study the traditional temperature-zero regret criterion, which evaluates only the top-ranked response at inference time. Under this decision-centric notion of performance, we prove that standard greedy online alignment methods, including online RLHF and online DPO, achieve constant \((O(1))\) cumulative regret. By isolating the cost of identifying the best response from the stochasticity induced by regularization, our results provide a sharper theoretical explanation for the practical superb efficiency of greedy alignment.

Enoch Hyunwook Kang

AI agents are increasingly deployed in interactive economic environments characterized by repeated AI-AI interactions. Despite AI agents' advanced capabilities, empirical studies reveal that such interactions often fail to stably induce a strategic equilibrium, such as a Nash equilibrium. Post-training methods have been proposed to induce a strategic equilibrium; however, it remains impractical to uniformly apply an alignment method across diverse, independently developed AI models in strategic settings. In this paper, we provide theoretical and empirical evidence that off-the-shelf reasoning AI agents can achieve Nash-like play zero-shot, without explicit post-training. Specifically, we prove that `reasonably reasoning' agents, i.e., agents capable of forming beliefs about others' strategies from previous observation and learning to best respond to these beliefs, eventually behave along almost every realized play path in a way that is weakly close to a Nash equilibrium of the continuation game. In addition, we relax the common-knowledge payoff assumption by allowing stage payoffs to be unknown and by having each agent observe only its own privately realized stochastic payoffs, and we show that we can still achieve the same on-path Nash convergence guarantee. We then empirically validate the proposed theories by simulating five game scenarios, ranging from a repeated prisoner's dilemma game to stylized repeated marketing promotion games. Our findings suggest that AI agents naturally exhibit such reasoning patterns and therefore attain stable equilibrium behaviors intrinsically, obviating the need for universal alignment procedures in many real-world strategic interactions.

Enoch Hyunwook Kang

Field experiments (A/B tests) are often the most credible benchmark for methods (algorithms) in societal systems, but their cost and latency bottleneck rapid methodological progress. LLM-based persona simulation offers a cheap synthetic alternative, yet it is unclear whether replacing humans with personas preserves the benchmark interface that adaptive methods optimize against. We prove an if-and-only-if characterization: when (i) methods observe only the aggregate outcome (aggregate-only observation) and (ii) evaluation depends only on the submitted artifact and not on the method's identity or provenance (method-blind evaluation), swapping humans for personas is just panel change from the method's point of view, indistinguishable from changing the evaluation population (e.g., New York to Jakarta). Furthermore, we move from validity to usefulness: we define an information-theoretic discriminability of the induced aggregate channel and show that making persona benchmarking as decision-relevant as a field experiment is fundamentally a sample-size question, yielding explicit bounds on the number of independent persona evaluations required to reliably distinguish meaningfully different methods at a chosen resolution.