IJCAI 2024 Papers

Auction-based Federated Learning (AFL) is a burgeoning research area. However, existing bidding strategies for AFL data consumers (DCs) primarily focus on maximizing expected accumulated utility, disregarding the more complex goal of revenue maximization. They also only consider winning bids, leading to biased estimates by overlooking information from losing bids. To address these issues, we propose a Bias-free Revenue-maximizing Federated bidding strategy for DCs in AFL (BR-FEDBIDDER). Our theoretical exploration of the relationships between Return on Investment (ROI), bid costs, and utility, and their impact on overall revenue underscores the complexity of maximizing revenue solely by prioritizing ROI enhancement. Leveraging these insights, BR-FEDBIDDER optimizes bid costs with any given ROI constraint. In addition, we incorporate an auxiliary task of winning probability estimation into the framework to achieve bias-free learning by leveraging bid records from historical bid requests, including both winning and losing ones. Extensive experiments on six widely used benchmark datasets show that BR-FEDBIDDER outperforms eight state-of-the-art methods, surpassing the best-performing baseline by 5.66%, 6.08% and 2.44% in terms of the total revenue, ROI, and test accuracy of the resulting FL models, respectively.

The bipartite traveling tournament problem (BTTP) was initially introduced by Hoshino and Kawarabayashi (AAAI 2011) to address inter-league sports scheduling, which aims to design a feasible bipartite tournament between two n-team leagues under some constraints such that the total traveling distance of all participating teams is minimized. Since its introduction, several heuristic methods have been developed to design feasible schedules for NBA, NPB and so on. In terms of solution quality with a theoretical guarantee, previously only a (2+ε) approximation is known for the case that n≡0 (mod 3). Whether there are similar results for the cases that n≡1 (mod 3) and n≡2 (mod 3) was asked in the literature. In this paper, we answer this question positively by proposing a (3/2+ε)-approximation algorithm for any n and any constant ε>0, which also improves the previous ratio for the case that n≡0 (mod 3).

Deep reinforcement learning is used in various domains, but usually under the assumption that the environment has stationary conditions like transitions and state distributions. When this assumption is not met, performance suffers. For this reason, tracking continuous environmental changes and adapting to unpredictable conditions is challenging yet crucial because it ensures that systems remain reliable and flexible in practical scenarios. Our research introduces Behavior-Aware Detection and Adaptation (BADA), an innovative framework that merges environmental change detection with behavior adaptation. The key inspiration behind our method is that policies exhibit different global behaviors in changing environments. Specifically, environmental changes are identified by analyzing variations between behaviors using Wasserstein distances without manually set thresholds. The model adapts to the new environment through behavior regularization based on the extent of changes. The results of a series of experiments demonstrate better performance relative to several current algorithms. This research also indicates significant potential for tackling this long-standing challenge.

Evaluating the performance of Multi-modal Large Language Models (MLLMs), integrating both point cloud and language, presents significant challenges. The lack of a comprehensive assessment hampers determining whether these models truly represent advancements, thereby impeding further progress in the field. Current evaluations heavily rely on classification and caption tasks, falling short in providing a thorough assessment of MLLMs. A pressing need exists for a more sophisticated evaluation method capable of thoroughly analyzing the spatial understanding and expressive capabilities of these models. To address these issues, we introduce a scalable 3D benchmark, accompanied by a large-scale instruction-tuning dataset known as 3DBench, providing an extensible platform for a comprehensive evaluation of MLLMs. Specifically, we establish the benchmark that spans a wide range of spatial and semantic scales, from object-level to scene-level, addressing both perception and planning tasks. Furthermore, we present a rigorous pipeline for automatically constructing scalable 3D instruction-tuning datasets, covering 10 diverse multi-modal tasks with more than 0.23 million QA pairs generated in total. Thorough experiments evaluating trending MLLMs, comparisons against existing datasets, and variations of training protocols demonstrate the superiority of 3DBench, offering valuable insights into current limitations and potential research directions. Codes are available at https://github.com/Inshsang/3DBench.

Molecular graph representation learning plays a crucial role in various domains, such as drug discovery and chemical reaction prediction, where molecular graphs are typically depicted as 2D topological structures. However, recent insights highlight the critical role of 3D geometric information and functional groups in accurately predicting molecular properties, aspects often neglected in existing molecular graph benchmark datasets. To bridge the research gap, we introduce a comprehensive molecular learning benchmark named 3D-FUM, which incorporates both 3D geometric information and functional groups of a large number of molecules. 3D-FUM integrates 18 state-of-the-art algorithms and 19 evaluation metrics on three molecular learning tasks, including general molecule generation, conditional molecule generation, and property predictions. 3D-FUM, for the first time, take into consideration both 3D geometric information and molecular functional groups, which enables researchers and practitioners to effectively and impartially evaluate newly proposed methods in comparison to existing baselines across diverse datasets. Furthermore, we design a user interface for user-friendly interaction and development with the benchmark for evaluation metrics selection, parameter adjustment, and leaderboard comparison. To ensure accessibility and reproducibility, we opensource our benchmark 3D-FUM and experimental results at https://3dfunctiongroupmoleculedataset.github.io/3D-FuM/#/Home.

3D Vision-Language Pre-training (3D-VLP) aims to provide a pre-train model which can bridge 3D scenes with natural language, which is an important technique for embodied intelligence. However, current 3D-VLP datasets are hindered by limited scene-level diversity and insufficient fine-grained annotations (only 1.2K scenes and 280K textual annotations in ScanScribe), primarily due to the labor-intensive of collecting and annotating 3D scenes. To overcome these obstacles, we construct SynVL3D, a comprehensive synthetic scene-text corpus with 10K indoor scenes and 1M descriptions at object, view, and room levels, which has the advantages of diverse scene data, rich textual descriptions, multi-grained 3D-text associations, and low collection cost. Utilizing the rich annotations in SynVL3D, we pre-train a simple and unified Transformer for aligning 3D and language with multi-grained pretraining tasks. Moreover, we propose a synthetic-to-real domain adaptation in downstream task fine-tuning process to address the domain shift. Through extensive experiments, we verify the effectiveness of our model design by achieving state-of-the-art performance on downstream tasks including visual grounding, dense captioning, and question answering. Codes are available at: https://github.com/idejie/3DSyn

Recently, heatmap regression methods based on 1D landmark representations have shown prominent performance on locating facial landmarks. However, previous methods ignored to make deep explorations on the good potentials of 1D landmark representations for sequential and structural modeling of multiple landmarks to track facial landmarks. To address this limitation, we propose a Transformer architecture, namely 1DFormer, which learns informative 1D landmark representations by capturing the dynamic and the geometric patterns of landmarks via token communications in both temporal and spatial dimensions for facial landmark tracking. For temporal modeling, we propose a confidence-enhanced multi-head attention mechanism with a recurrently token mixing strategy to adaptively and robustly embed long-term landmark dynamics into their 1D representations; for structure modeling, we design intra-group and inter-group geometric encoding mechanisms to encode the component-level as well as global-level facial structure patterns as a refinement for the 1D representations of landmarks through token communications in the spatial dimension via 1D convolutional layers. Experimental results on the 300VW and the TF databases show that 1DFormer successfully models the long-range sequential patterns as well as the inherent facial structures to learn informative 1D representations of landmark sequences, and achieves state-of-the-art performance on facial landmark tracking. Codes of our model are available in the supplementary materials.