United States

In this paper, we delve into the utilization of the negative momentum technique in constrained minimax games. From an intuitive mechanical standpoint, we introduce a novel framework for momentum buffer updating, which extends the findings of negative momentum from the unconstrained setting to the constrained setting and provides a universal enhancement to the classic game-solver algorithms. Additionally, we provide theoretical guarantees of convergence for our momentum-augmented learning algorithms. We then extend these algorithms to their extensive-form counterparts. Experimental results on both Normal Form Games (NFGs) and Extensive Form Games (EFGs) demonstrate that our momentum techniques can significantly improve algorithm performance, surpassing both their original versions and the SOTA baselines by a large margin.

AAAI 2025

Rapid Learning in Constrained Minimax Games with Negative Momentum

reinforcement learning

technical paper

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

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-25 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.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

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-25 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.



Live comments, also known as Danmaku, are user-generated messages that are synchronized with video content. These comments overlay directly onto streaming videos, capturing viewer emotions and reactions in real-time. While prior work has leveraged live comments in affective analysis, its use has been limited due to the relative rarity of live comments across different video platforms. To address this, we first construct the Live Comment for Affective Analysis (LCAffect) dataset, which contains live comments for English and Chinese videos spanning diverse genres that elicit a wide spectrum of emotions. Then, using this dataset, we use contrastive learning to train a video encoder to produce synthetic live comment features for enhanced multimodal affective content analysis. Through comprehensive experimentation on a wide range of affective analysis tasks (sentiment, emotion recognition, and sarcasm detection) in both English and Chinese, we demonstrate that these synthetic live comment features significantly improve performance over state-of-the-art methods. We will release our dataset, code, and models upon publication.

Enhancing Multimodal Affective Analysis with Learned Live Comment Features

Fine-tuning text-to-image diffusion models to maximize rewards has proven effective for enhancing model performance. However, reward fine-tuning methods often suffer from slow convergence due to online sample generation. Therefore, obtaining diverse samples with strong reward signals is crucial for improving sample efficiency and overall performance. In this work, we introduce DiffExp, a simple yet effective exploration strategy for reward fine-tuning of of text-to-image models. Our approach employs two key strategies: (a) dynamically adjusting the scale of classifier-free guidance to enhance sample diversity, and (b) randomly weighting phrases of the text prompt to exploit high-quality reward signals. We demonstrate that these strategies significantly enhance exploration during online sample generation, improving the sample efficiency of recent reward fine-tuning methods, such as DDPO and AlignProp.

DiffExp: Efficient Exploration in Reward Fine-tuning for Text-to-Image Diffusion Models

Diffusion-based molecular graph generative models have achieved significant success in template-free, single-step retrosynthesis prediction. However, these models typically generate reactants from scratch, often overlooking the fact that the scaffold of a product molecule typically remains unchanged during chemical reactions. To leverage this useful observation, we introduce a retrieval-augmented molecular graph generation framework. Our framework comprises three key components: a retrieval component that identifies similar molecules for the given product, an integration component that learns valuable clues from these molecules about which part of the product should remain unchanged, and a base generative model that is prompted by these clues to generate the corresponding reactants. We explore various design choices for critical and under-explored aspects of this framework and instantiate it as the Retrieval-Augmented RetroBridge (RARB). RARB demonstrates state-of-the-art performance on standard benchmarks, achieving a 14.8% relative improvement in top-1 accuracy over its base generative model, highlighting the effectiveness of retrieval augmentation. Additionally, RARB excels in handling out-of-distribution molecules, and its advantages remain significant even with smaller models or fewer denoising steps. These strengths make RARB highly valuable for real-world retrosynthesis applications, where extrapolation to novel molecules and high-throughput prediction are essential. Our code is available at this repository: https://anonymous.4open.science/r/RARB-23B3.

Advancing Retrosynthesis with Retrieval-Augmented Graph Generation

We formulate the problem of fair and efficient completion of indivisible goods, defined as follows: Given a partial allocation of indivisible goods among agents, does there exist an allocation of the remaining goods (i.e., a completion) that satisfies fairness and economic efficiency guarantees of interest? We study the computational complexity of the completion problem for prominent fairness and efficiency notions such as envy-freeness up to one good (EF1), proportionality up to one good (Prop1), maximin share (MMS), and Pareto optimality (PO), and focus on the class of additive valuations as well as its subclasses such as binary additive and lexicographic valuations. We find that while the completion problem is significantly harder than the standard fair division problem (wherein the initial partial allocation is empty), the consideration of restricted preferences facilitates positive algorithmic results for threshold-based fairness notions (Prop1 and MMS). On the other hand, the completion problem remains computationally intractable for envy-based notions such as EF1 and EF1+PO even under restricted preferences.

Fair and Efficient Completion of Indivisible Goods

Knowledge distillation (KD) has become a prevalent technique for compressing large language models (LLMs). Existing KD methods are constrained by the need for identical tokenizers (i.e., vocabularies) between teacher and student models, limiting their versatility in handling LLMs of different architecture families. In this paper, we introduce the Multi-Level Optimal Transport (MultiLevelOT), a novel approach that advances the optimal transport for universal cross-tokenizer knowledge distillation. Our method aligns the logit distributions of the teacher and the student at both token and sequence levels using diverse cost matrices, eliminating the need for dimensional or token-by-token correspondence. At the token level, MultiLevelOT integrates both global and local information by jointly optimizing all tokens within a sequence to enhance robustness. At the sequence level, we efficiently capture complex distribution structures of logits via the Sinkhorn distance, which approximates the Wasserstein distance for divergence measures. Extensive experiments on tasks such as extractive QA, generative QA, and summarization demonstrate that the MultiLevelOT outperforms state-of-the-art cross-tokenizer KD methods under various settings.
Our approach is robust to different student and teacher models across model families, architectures, and parameter sizes. Codes and models are available at https://anonymous.4open.science/r/AAAI2025-1450.

Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models

Image super-resolution (SR) is essential for bridging the gap between modern hardware and real-time computer graphics (CG) applications. It reduces CG workload by allowing low-resolution rendering, with original quality restored later via mathematical operations or machine learning. However, recent learning-based SR methods often rely on complex models, demanding high computational resources and undermining the benefits of reduced rendering workload.

Our qualitative and quantitative analysis of the SR process and rendering reveals that readily accessible rendering information can significantly enhance neural network design by serving as additional features. To capitalize on this, we propose CGSR, an optimization framework designed for lightweight real-time super-resolution. CGSR utilizes rendering information to boost both network extensibility and efficiency. It utilizes progressively available rendering information from the pipeline, which arrives earlier than the rendered frame, enabling pre-processing and masking of latency. These features are then integrated into a selected SR network backbone to form a CG-enhanced network. This network is further optimized and refined into a CG-optimized version using neural architecture search (NAS). To improve runtime performance, CGSR also employs rendering-aware hybrid pruning, which dynamically prunes the network based on temporal rendering data. Evaluation results show that CGSR significantly reduces parameter size, multi-add operations, and inference time while maintaining high SR quality across various backbone SR networks.

Achieving Lightweight Super-Resolution for Real-Time Computer Graphics

Humans achieve contact-rich dexterous grasping through the synergy of visual and tactile information. However, the high-dimensional action space of high DoF multi-fingered hands poses significant challenges to this operation. In this study, we address this complexity by controlling the robotic hand at the reduced dimensional level of individual fingers instead of the entire hand, and develop a finger-based multi-agent deep reinforcement learning strategy by regarding the wrist, arm, and each finger of the hand as intelligent agents. We commence by applying a single-agent reinforcement learning algorithm to guide the whole hand to reach the feasible approaching direction and distance to the object. Then, we develop neuroscience-inspired visuo-tactile fusion networks to train multiple agents to control their assigned fingers by effectively leveraging visual and tactile feedback. This enables dynamic and collaborative adjustments of finger-object interactions, ultimately achieving precise contact with specific areas of the objects. The grasping results on 8 objects show that our approach can achieve stable and compliant grasps. To the best of our knowledge, this is the first work that employs a finger-based multi-agent reinforcement learning approach to control the dexterous grasping process under the guidance of both visual and tactile feedback.

Multi-fingered Hand Grasps with Visuo-Tactile Fusion via Multi-Agent Deep Reinforcement Learning

Talking face generation (TFG), a pivotal technology for digital human creation, allows for producing lifelike talking videos of any character using only facial images and accompanying text. Despite the urgent necessity for targeted detection methods, research in this field has been hindered by the lack of public datasets. In this paper, we construct the first large-scale multi-scenario talking face dataset (MSTF), which contains 22 audio and video forgery techniques, filling the gap of datasets in this field. The dataset covers 11 generation scenarios and more than 20 semantic scenarios, closer to the practical application scenario of TFG. Besides, we also propose a TFG detection framework, which leverages the analysis of both global and local coherence in the multimodal content of TFG videos. Therefore, a region-focused smoothness detection module (RSFDM) and a discrepancy capture-time frame aggregation module (DCTAM) are introduced to evaluate the global temporal coherence of TFG videos, aggregating multi-grained spatial information. Additionally, a visual-audio fusion module (V-AFM) is designed to evaluate audiovisual coherence within a localized temporal perspective. Comprehensive experiments demonstrate the reasonableness and challenges of our datasets, while also indicating the superiority of our proposed method compared to the state-of-the-art deepfake detection approaches.

GLCF: A Global-Local Multimodal Coherence Analysis Framework for Talking Face Generation Detection

Deep reinforcement learning (DRL) has achieved remarkable success in various domains, yet its reliance on neural networks results in a lack of transparency, which limits its practical applications in safety-critical and human-agent interaction domains. Decision trees, known for their notable explainability, have emerged as a promising alternative to neural networks. However, decision trees often struggle in long-horizon continuous control tasks with high-dimensional observation space due to their limited expressiveness. To address this challenge, we propose SkillTree, a novel hierarchical framework that reduces the complex continuous action space of challenging control tasks into discrete skill space. By integrating the differentiable decision tree within the high-level policy, SkillTree generates diecrete skill embeddings that guide low-level policy execution. Furthermore, through distillation, we obtain a simplified decision tree model that improves performance while further reducing complexity. Experiment results validate SkillTree's effectiveness across various robotic manipulation tasks, providing clear skill-level insights into the decision-making process. The proposed approach not only achieves performance comparable to neural network based methods in complex long-horizon control tasks but also significantly enhances the transparency and explainability of the decision-making process.

SkillTree: Explainable Skill-Based Deep Reinforcement Learning for Long-Horizon Control Tasks

Time Series Supplier Allocation (TSSA) poses a complex NP-hard challenge, aimed at refining future order dispatching strategies to satisfy order demands with maximum supply efficiency fully. 
Traditionally derived from financial portfolio management, the Black-Litterman (BL) model offers a new perspective for the TSSA scenario by balancing expected returns against insufficient supply risks. However, its application within TSSA is constrained by the reliance on manually constructed perspective matrices and spatio-temporal market dynamics, coupled with the absence of supervisory signals and data unreliability inherent to supplier information.
To solve these limitations, we introduce the pioneering Deep Black-Litterman Model (DBLM), which innovatively adapts the BL model from financial roots to supply chain context. Leveraging the Spatio-Temporal Graph Neural Networks (STGNNs), DBLM automatically generates future perspective matrices for TSSA, by integrating spatio-temporal dependency. 
Moreover, a novel Spearman rank correlation distinctively supervises our approach to address the lack of supervisory signals, specifically designed to navigate through the complexities of supplier risks and interactions.
This is further enhanced by a masking mechanism aimed at counteracting the biases from unreliable data, thereby improving the model’s precision and reliability. Extensive experimentation on two datasets unequivocally demonstrates DBLM's~ enhanced performance in TSSA, setting new standards for the field. Our findings and methodology are made available for community access and further development at https://anonymous.4open.science/r/DBLM-7978/.

Rapid Learning in Constrained Minimax Games with Negative Momentum

Downloads

Next from AAAI 2025

Enhancing Multimodal Affective Analysis with Learned Live Comment Features

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES