We propose Constraint-Generation Policy Optimization (CGPO) for optimizing policy parameters within compact and interpretable policy classes for mixed discrete-continuous Markov Decision Processes (DC-MDPs). CGPO is not only able to provide bounded policy error guarantees over an infinite range of initial states for many DC-MDPs with expressive nonlinear dynamics, but it can also provably derive optimal policies in cases where it terminates with zero error. Furthermore, CGPO can generate worst-case state trajectories to diagnose policy deficiencies and provide counterfactual explanations of optimal actions. To achieve such results, CGPO proposes a bi-level mixed-integer nonlinear optimization framework for optimizing policies within defined expressivity classes (i.e. piecewise (non)-linear) and reduces it to an optimal constraint generation methodology that adversarially generates worst-case state trajectories. Furthermore, leveraging modern nonlinear optimizers, CGPO can obtain solutions with bounded optimality gap guarantees. We handle stochastic transitions through explicit marginalization (where applicable) or chance-constraints, providing high-probability policy performance guarantees. We also present a road-map for understanding the computational complexities associated with different expressivity classes of policy, reward, and transition dynamics. We experimentally demonstrate the applicability of CGPO in diverse domains, including inventory control, management of a system of water reservoirs, and physics control. In summary, we provide a solution for deriving structured, compact, and explainable policies with bounded performance guarantees, enabling worst-case scenario generation and counterfactual policy diagnostics.

Diffusion models learn to reverse the progressive noising of a data distribution to create a generative model. However, the desired continuous nature of the noising process can be at odds with discrete data. To deal with this tension between continuous and discrete objects, we propose a method of performing diffusion on the probability simplex. Using the probability simplex naturally creates an interpretation where points correspond to categorical probability distributions. Our method uses the softmax function applied to an Ornstein-Unlenbeck Process, a well-known stochastic differential equation. We find that our methodology also naturally extends to include diffusion on the unit cube which has applications for bounded image generation.

As natural language interfaces enable users to express increasingly complex natural language queries, there is a parallel explosion of user review content that can allow users to better find items such as restaurants, books, or movies that match these expressive queries. While Neural Information Retrieval (IR) methods have provided state-of-theart results for matching queries to documents, they have not been extended to the task of Reviewed-Item Retrieval (RIR), where query-review scores must be aggregated (or fused) into item-level scores for ranking. In the absence of labeled RIR datasets, we extend Neural IR methodology to RIR by leveraging self-supervised methods for contrastive learning of BERT embeddings for both queries and reviews. Specifically, contrastive learning requires a choice of positive and negative samples, where the unique two-level structure of our item-review data combined with metadata affords us a rich structure for the selection of these samples. For contrastive learning in a Late Fusion scenario (where we aggregate queryreview scores into item-level scores), we investigate the use of positive review samples from the same item and/or with the same rating, selection of hard positive samples by choosing the least similar reviews from the same anchor item, and selection of hard negative samples by choosing the most similar reviews from different items. We also explore anchor sub-sampling and augmenting with meta-data. For a more end-to-end Early Fusion approach, we introduce contrastive item embedding learning to fuse reviews into single item embeddings. Experimental results show that Late Fusion contrastive learning for Neural RIR outperforms all other contrastive IR configurations, Neural IR, and sparse retrieval baselines, thus demonstrating the power of exploiting the two-level structure in Neural RIR approaches as well as the importance of preserving the nuance of individual review content via Late Fusion methods.

Traditional recommender systems leverage users' item preference history to recommend novel content that users may like. However, modern dialog interfaces that allow users to express language-based preferences offer a fundamentally different modality for preference input. Inspired by recent successes of prompting paradigms for large language models (LLMs), we study their use for making recommendations from both item-based and language-based preferences in comparison to state-of-the-art item-based collaborative filtering (CF) methods. To support this investigation, we collect a new dataset consisting of both item-based and language-based preferences elicited from users along with their ratings on a variety of (biased) recommended items and (unbiased) random items. Among numerous experimental results, we find that LLMs provide competitive recommendation performance for pure language-based preferences (no item preferences) in the near cold-start case in comparison to item-based CF methods, despite having no supervised training for this specific task (zero-shot) or only a few labels (few-shot). This is particularly promising as language-based preference representations are more explainable and scrutable than item-based or vector-based representations.

Reinforcement Learning (RL) Sutton and Barto [2018] and Probabilistic planning Puterman [2014] are two research branches that address stochastic problems, often under the Markov assumption for state dynamics. The planning approach requires a given model, while the learning approach improves through repeated interaction with an environment, which can be viewed as a black box. Thus, the tools and the benchmarks for these two branches have grown apart. Learning agents do not require to be able to simulate model-based transitions, and thus frameworks such as OpenAI Gym Brockman et al. [2016] have become a standard, serving also as an interface for third-party benchmarks such as Todorov et al. [2012], Bellemare et al. [2013] and more. As the model is not necessary for solving the learning problem, the environments are hard-coded in a programming language. This has several downsides; if one does wish to see the model describing the environment, it has to be reverse-engineered from the environment framework, complex problems can result in a significant development period, code bugs may make their way into the environment and finally, there is no clean way to verify the model or reuse it directly. Thus, the creation of a verified acceptable benchmark is a challenging task. Planning agents on the other hand can interact with an environment Sanner [2010a], but in many cases simulate the model within the planning agent in order to solve the problem Keller and Eyerich [2012]. The planning community has also come up with formal description languages for various types of problems; these include the Planning Domain Definition Language (PDDL) Aeronautiques et al. [1998] for classical planning problems, PDDL2.1 Fox and Long [2003] for problems involving time and continuous variables, PPDDL Bryce and Buet [2008] for classical planning problems with action probabilistic effects and rewards, and Relational Dynamic Influence Diagram Language (RDDL)

Text detoxification is a conditional text generation task aiming to remove offensive content from toxic text. It is highly useful for online forums and social media, where offensive content is frequently encountered. Intuitively, there are diverse ways to detoxify sentences while preserving their meanings, and we can select from detoxified sentences before displaying text to users. Conditional diffusion models are particularly suitable for this task given their demonstrated higher generative diversity than existing conditional text generation models based on language models. Nonetheless, text fluency declines when they are trained with insufficient data, which is the case for this task. In this work, we propose DiffuDetox, a mixed conditional and unconditional diffusion model for text detoxification. The conditional model takes toxic text as the condition and reduces its toxicity, yielding a diverse set of detoxified sentences. The unconditional model is trained to recover the input text, which allows the introduction of additional fluent text for training and thus ensures text fluency. Extensive experimental results and in-depth analysis demonstrate the effectiveness of our proposed DiffuDetox.

Perimeter control maintains high traffic efficiency within protected regions by controlling transfer flows among regions to ensure that their traffic densities are below critical values. Existing approaches can be categorized as either model-based or model-free, depending on whether they rely on network transmission models (NTMs) and macroscopic fundamental diagrams (MFDs). Although model-based approaches are more data efficient and have performance guarantees, they are inherently prone to model bias and inaccuracy. For example, NTMs often become imprecise for a large number of protected regions, and MFDs can exhibit scatter and hysteresis that are not captured in existing model-based works. Moreover, no existing studies have employed reinforcement learning for homogeneous flow rate optimization in microscopic simulation, where spatial characteristics, vehicle-level information, and metering realizations -- often overlooked in macroscopic simulations -- are taken into account. To circumvent issues of model-based approaches and macroscopic simulation, we propose a model-free deep reinforcement learning approach that optimizes the flow rate homogeneously at the perimeter at the microscopic level. Results demonstrate that our model-free reinforcement learning approach without any knowledge of NTMs or MFDs can compete and match the performance of a model-based approach, and exhibits enhanced generalizability and scalability.

Traffic prediction is a spatiotemporal predictive task that plays an essential role in intelligent transportation systems. Today, graph convolutional neural networks (GCNNs) have become the prevailing models in the traffic prediction literature since they excel at extracting spatial correlations. In this work, we classify the components of successful GCNN prediction models and analyze the effects of matrix factorization, attention mechanism, and weight sharing on their performance. Furthermore, we compare these variations against random forests, a traditional regression method that predates GCNNs by over 15 years. We evaluated these methods using simulated data of two regions in Toronto as well as real-world sensor data from selected California highways. We found that incorporating matrix factorization, attention, and location-specific model weights either individually or collectively into GCNNs can result in a better overall performance. Moreover, although random forest regression is a less compact model, it matches or exceeds the performance of all variations of GCNNs in our experiments. This suggests that the current graph convolutional methods may not be the best approach to traffic prediction and there is still room for improvement. Finally, our findings also suggest that for future research on GCNN for traffic prediction to be credible, researchers must include performance comparison to random forests.

Can a Large Language Model (LLM) solve simple abstract reasoning problems? We explore this broad question through a systematic analysis of GPT on the Abstraction and Reasoning Corpus (ARC) [5], a representative benchmark of abstract reasoning ability from limited examples in which solutions require some "core knowledge" of concepts such as objects, goal states, counting, and basic geometry. GPT-4 solves only 13/50 of the most straightforward ARC tasks when using textual encodings for their two-dimensional input-output grids. Our failure analysis reveals that GPT-4's capacity to identify objects and reason about them is significantly influenced by the sequential nature of the text that represents an object within a text encoding of a task. To test this hypothesis, we design a new benchmark, the 1D-ARC, which consists of one-dimensional (array-like) tasks that are more conducive to GPT-based reasoning, and where it indeed performs better than on the (2D) ARC. To alleviate this issue, we propose an object-based representation that is obtained through an external tool, resulting in nearly doubling the performance on solved ARC tasks and near-perfect scores on the easier 1D-ARC. Although the state-of-the-art GPT-4 is unable to "reason" perfectly within non-language domains such as the 1D-ARC or a simple ARC subset, our study reveals that the use of object-based representations can significantly improve its reasoning ability.

Clustering is a powerful and extensively used data science tool. While clustering is generally thought of as an unsupervised learning technique, there are also supervised variations such as Spath's clusterwise regression that attempt to find clusters of data that yield low regression error on a supervised target. We believe that clusterwise regression is just a single vertex of a largely unexplored design space of supervised clustering models. In this article, we define a generalized optimization framework for predictive clustering that admits different cluster definitions (arbitrary point assignment, closest center, and bounding box) and both regression and classification objectives. We then present a joint optimization strategy that exploits mixed-integer linear programming (MILP) for global optimization in this generalized framework. To alleviate scalability concerns for large datasets, we also provide highly scalable greedy algorithms inspired by the Majorization-Minimization (MM) framework. Finally, we demonstrate the ability of our models to uncover different interpretable discrete cluster structures in data by experimenting with four real-world datasets.