United States

The policy-based reinforcement learning (RL) can be considered as maximization of its objective. However, due to the inherent non-concavity of its objective, the policy gradient method to a first-order stationary point (FOSP) cannot guar- antee a maximal point. A FOSP can be a minimal or even a saddle point, which is undesirable for RL. It has be found that if all the saddle points are strict, all the second-order station- ary points (SOSP) are exactly equivalent to local maxima. Instead of FOSP, we consider SOSP as the convergence criteria to characterize the sample complexity of policy gradient. Our result shows that policy gradient converges to an (Îµ, âˆšÎµÏ‡)-SOSP with probability at least 1 âˆ’ O(Î´) after the total cost of O(Îµâˆ’9/2)sinificantly improves the state of the art cost O(Îµâˆ’9).Our analysis is based on the key idea that decomposes the parameter space Rp into three non-intersected regions: non-stationary point region, saddle point region, and local optimal region, then making a local improvement of the objective of RL in each region. This technique can be potentially generalized to extensive policy gradient methods. For the complete proof, please refer to https://arxiv.org/pdf/2012.01491.pdf.

AAAI 2021

Sample Complexity of Policy Gradient Finding Second-Order Stationary Points

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-26 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.<br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to AAAI 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://aaai.org/conference/aaai/aaai-26/registration/) first.

**Online Registration Form**: https://aaai.getregistered.net/conference-2026 

Registration Required

AAAI 2026

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

technical paper

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

Like humans, document summarization models can interpret a documentâ€™s contents in a number of ways. Unfortunately, the  neural models of today are largely black boxes that provide little explanation of how or why they generated a summary in the way they did. Therefore, to begin prying open the black box and to inject a level of control into the substance of the final summary, we developed a novel select-and-generate framework that focuses on explainability. By revealing the latent centrality and interactions between sentences, along with scores for novelty and relevance, users are given a window into the choices a model is making and an opportunity to guide those choices in a more desirable direction. A novel pair-wise matrix captures the sentence interactions, centrality and attribute scores, and a mask with tunable attribute thresholds allows the user to control which sentences are likely to be included in the extraction. A sentence-deployed attention mechanism in the abstractor ensures the final summary emphasizes the desired content. Additionally, the encoder is adaptable, supporting both Transformer- and BERT-based configurations. In a series of experiments assessed with ROUGE metrics and two human evaluations, ESCA outperformed eight state-of-the-art models on the CNN/DailyMail and NYT50 benchmark datasets.

Exploring Explainable Selection to Control Abstractive Summarization

One of the key factors of enabling machine learning models to comprehend and solve real-world tasks is to leverage multimodal data. Unfortunately, annotation of multimodal data is challenging and expensive. Recently, self-supervised multimodal methods that combine vision and language were proposed to learn multimodal representations without annotation. However, these methods often choose to ignore the presence of high levels of noise and thus yield sub-optimal results. In this work, we show that the problem of noise estimation for multimodal data can be reduced to a multimodal density estimation task. Using multimodal density estimation, we propose a noise estimation building block for multimodal representation learning that is based strictly on the inherent correlation between different modalities. We demonstrate how our noise estimation can be broadly integrated and achieves comparable results to state-of-the-art performance on five different benchmark datasets for two challenging multimodal tasks: Video Question Answering and Text-To-Video Retrieval. Furthermore, we provide a theoretical probabilistic error bound substantiating our empirical results and analyze failure cases. Code: https://github.com/elad-amrani/ssml.

Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning

We present a new approach for learning the structure of a treewidth-bounded Bayesian Network (BN). The key to our approach is applying an exact method (based on MaxSAT) locally, to improve the score of a heuristically computed BN. This approach allows us to scale the power of exact methodsâ€”so far only applicable to BNs with several dozens of random variablesâ€”to large BNs with several thousands of random variables. Our experiments show that our method improves the score of BNs provided by state-of-the-art heuristic methods, often significantly.

Turbocharging Treewidth-Bounded Bayesian Network Structure Learning

The robustness of neural networks can be quantitatively indicated by a lower bound within which any perturbation does not alter the original inputâ€™s classification result. A certified lower bound is also a criterion to evaluate the performance of robustness verification approaches. In this paper, we present a tighter linear approximation approach for the robustness verification of Convolutional Neural Networks (CNNs). By the tighter approximation, we can tighten the robustness verification of CNNs, i.e., proving they are robust within a larger 10 perturbation distance. Furthermore, our approach is applicable to general sigmoid-like activation functions. We implement DeepCert, the resulting verification toolkit. We evaluate it with open-source benchmarks, including LeNet and the models trained on MNIST and CIFAR. Experimental results show that DeepCert outperforms other state-of-the-art robustness verification tools with at most 286.28% improvement to the certified lower bound and 1566.76 times speedup for the same neural networks.

Tightening Robustness Verification of Convolutional Neural Networks with Fine-Grained Linear Approximation

Stochastic Boolean Satisfiability (SSAT) is a powerful representation for the concise encoding of quantified decision problems with uncertainty. While it shares commonalities with quantified Boolean formula (QBF) satisfiability and has the same PSPACE-complete complexity, SSAT solving tends to be more challenging as it involves expensive model counting, a.k.a. Sharp-SAT. To date, SSAT solvers, especially those imposing no restrictions on quantification levels, remain much lacking. In this paper, we present a new SSAT solver based on the framework of clause selection and cube distribution previously proposed for QBF solving. With model counting integrated and learning techniques strengthened, our solver is general and effective. Experimental results demonstrate the overall superiority of the proposed algorithm in both solving performance and memory usage compared to the state-of-the-art solvers on a number of benchmark formulas.

A Sharp Leap from Quantified Boolean Formula to Stochastic Boolean Satisfiability Solving

At the working heart of policy iteration algorithms commonly used and studied in the discounted setting of reinforcement learning, the policy evaluation step estimates the value of states with samples from a Markov reward process induced by following a Markov policy in a Markov decision process. We propose a simple and efficient estimator called loop estimator that exploits the regenerative structure of Markov reward processes without explicitly estimating a full model. Our method enjoys a space complexity of O(1) when estimating the value of a single positive recurrent state s unlike TD with O(S) or model-based methods with O(S^2). Moreover, the regenerative structure enables us to show, without relying on the generative model approach, that the estimator has an instance-dependent convergence rate of O~(\sqrt{\tau_s/T}) over steps T on a single sample path, where \tau_s is the maximal expected hitting time to state s. In preliminary numerical experiments, the loop estimator outperforms model-free methods, such as TD(k), and is competitive with the model-based estimator.

Loop Estimator for Discounted Values in Markov Reward Processes

We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning system's dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equations, and give rise to improved performance on time-series prediction tasks. To demonstrate these properties, we first take a theoretical approach to find bounds over their dynamics, and compute their expressive power by the trajectory length measure in a latent trajectory space. We then conduct a series of time-series prediction experiments to manifest the approximation capability of Liquid Time-Constant Networks (LTCs) compared to classical and modern RNNs.

Liquid Time-Constant Networks

Multi-headed attention heads are a mainstay in transformer-based models. Different methods have been proposed to classify the role of each attention head based on the relations between tokens which have high pair-wise attention. These roles include syntactic (tokens with some syntactic relation), local (nearby tokens), block (tokens in the same sentence) and delimiter (the special [CLS], [SEP] tokens). There are two main challenges with existing methods for classification: (a) there  are no standard scores across studies or across functional roles, and (b) these scores are often average quantities measured across sentences without capturing statistical significance. In this work, we formalize a simple yet effective score that generalizes to all the roles of attention heads and employs hypothesis testing on this score for robust inference. This provides us the right lens to systematically analyze attention heads and confidently comment on many commonly posed questions on analyzing the BERT model. In particular, we comment on the co-location of multiple functional roles in the same attention head, the distribution of attention heads across layers, and effect of fine-tuning for specific NLP tasks on these functional roles. The code is made publicly available at https://github.com/iitmnlp/heads-hypothesis

The Heads Hypothesis: A Unifying Statistical Approach towards Understanding Multi-Headed Attention in BERT

End-to-end semantic role labeling (SRL) has been received increasing interest. It performs the two subtasks of SRL: predicate identification and argument role labeling, jointly. Recent work is mostly focused on graph-based neural models, while the transition-based framework with neural networks which has been widely used in a number of closely-related tasks, has not been studied for the joint task yet. In this paper, we present the first work of transition-based neural models for end-to-end SRL. Our transition model incrementally discovers all sentential predicates as well as their arguments by a set of transition actions. The actions of the two subtasks are executed mutually for full interactions. Besides, we suggest high-order compositions to extract non-local features, which can enhance the proposed transition model further. Experimental results on CoNLL09 and Universal Proposition Bank show that our final model can produce state-of-the-art performance, and meanwhile keeps highly efficient in decoding. We also conduct detailed experimental analysis for a deep understanding of our proposed model.

End-to-End Semantic Role Labeling with Neural Transition-Based Model

This paper studies the problem of learning interactive recommender systems from logged  feedbacks without any exploration in online environments. We address the problem by proposing a general offline reinforcement learning framework for recommendation, which enables maximizing cumulative user rewards without online exploration. Specifically, we first introduce a probabilistic generative model for  interactive recommendation, and then propose an effective inference algorithm for discrete and stochastic policy learning based on logged  feedbacks. In order to perform offline learning more effectively, we propose five approaches to minimize the distribution mismatch between the logging policy and recommendation policy: support constraints, supervised regularization, policy constraints, dual constraints and reward extrapolation. We conduct extensive  experiments on two public real-world datasets,  demonstrating that the proposed methods can achieve superior performance over existing supervised learning and reinforcement learning methods for  recommendation.

Downloads

Next from AAAI 2021

Exploring Explainable Selection to Control Abstractive Summarization

Similar lecture

Policy Newton Methods for Distortion Riskmetrics

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2021

Exploring Explainable Selection to Control Abstractive Summarization

Similar lecture

Policy Newton Methods for Distortion Riskmetrics

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads