Diffusion models (DMs) have attracted attention in generative modeling due to their ability to produce high-quality, diverse outputs by progressively adding noise to data and then denoising it. However, DMs are computationally intensive due to their iterative nature, requiring numerous forward passes and high-precision operations, making them less efficient for resource-constrained environments. Recent efforts to reduce these computational demands using quantization show promise by converting high-precision parameters to lower precision, but they face challenges unique to DMs, particularly in addressing cross-timestep error propagation in the iterative process. In this paper, we analyze cross-timestep error propagation in quantized DMs, revealing that previous methods focusing only on reducing noise estimation discrepancies are insufficient. Instead, we introduce Cross-Timestep Error Correction (CTEC), where the quantized model not only approximates the full-precision model but also corrects errors from the previous timestep. A distillation method is applied to learn this correction process effectively. We conduct extensive experiments on unconditional image generation with LSUN-Churches and LSUN-Bedrooms, as well as conditional image generation with ImageNet. Our findings demonstrate the effectiveness of our method in significantly reducing accumulated quantization errors across timesteps within the quantized diffusion process. This enhancement enables the generation of high-quality images, even when constrained by reduced bitwidths.
论文检索
输入标题、作者或关键词，从 11,272 篇学术成果中精准定位
Diffusion models have shown promising ability in generating high-quality time series (TS) data. Despite the initial success, existing works mostly focus on the authenticity of data at the individual level, but pay less attention to preserving the population-level properties on the entire dataset. Such population-level properties include value distributions for each dimension and distributions of certain functional dependencies (e.g., cross-correlation, CC) between different dimensions. For instance, when generating house energy consumption TS data, the value distributions of the outside temperature and the kitchen temperature should be preserved, as well as the distribution of CC between them. Preserving such TS population-level properties is critical in maintaining the statistical insights of the datasets, mitigating model bias, and augmenting downstream tasks like TS prediction. Yet, it is often overlooked by existing models. Hence, data generated by existing models often bear distribution shifts from the original data. We propose Population-aware Diffusion for Time Series (PaD-TS), a new TS generation model that better preserves the population-level properties. The key novelties of PaD-TS include 1) a new training method explicitly incorporating TS population-level property preservation, and 2) a new dual-channel encoder model architecture that better captures the TS data structure. Empirical results in major benchmark datasets show that PaD-TS can improve the average CC distribution shift score between real and synthetic data by 5.9x while maintaining a performance comparable to state-of-the-art models on individual-level authenticity.
Differential equations are widely used to describe complex dynamical systems with evolving parameters in nature and engineering. Effectively learning a family of maps from the parameter function to the system dynamics is of great significance. In this study, we propose a novel learning framework of symbolic continuous-depth neural networks, termed Symbolic Neural Ordinary Differential Equations (SNODEs), to effectively and accurately learn the underlying dynamics of complex systems. Specifically, our learning framework comprises three stages: initially, pre-training a predefined symbolic neural network via a gradient flow matching strategy; subsequently, fine-tuning this network using Neural ODEs; and finally, constructing a general neural network to capture residuals. In this process, we apply the SNODEs framework to partial differential equation systems through Fourier analysis, achieving resolution-invariant modeling. Moreover, this framework integrates the strengths of symbolism and connectionism, boasting a universal approximation theorem while significantly enhancing interpretability and extrapolation capabilities relative to state-of-the-art baseline methods. We demonstrate this through experiments on several representative complex systems. Therefore, our framework can be further applied to a wide range of scientific problems, such as system bifurcation and control, reconstruction and forecasting, as well as the discovery of new equations.
The core challenge of high-dimensional and expensive black-box optimization (BBO) is how to obtain better performance faster with little function evaluation cost. The essence of the problem is how to design an efficient optimization strategy tailored to the target task. This paper designs a powerful optimization framework to automatically learn the optimization strategies from the target or cheap surrogate task without human intervention. However, current methods are weak for this due to poor representation of optimization strategy. To achieve this, 1) drawing on the mechanism of genetic algorithm, we propose a deep neural network framework called B2Opt, which has a stronger representation of optimization strategies based on survival of the fittest; 2) B2Opt can utilize the cheap surrogate functions of the target task to guide the design of the efficient optimization strategies. Compared to the state-of-the-art BBO baselines, B2Opt can achieve multiple orders of magnitude performance improvement with less function evaluation cost.
The convergence behavior of Stochastic Gradient Descent (SGD) crucially depends on the stepsize configuration. When using a constant stepsize, the SGD iterates form a Markov chain, enjoying fast convergence during the initial transient phase. However, when reaching stationarity, the iterates oscillate around the optimum without making further progress. In this paper, we study the convergence diagnostics for SGD with constant stepsize, aiming to develop an effective dynamic stepsize scheme. We propose a novel coupling-based convergence diagnostic procedure, which monitors the distance of two coupled SGD iterates for stationarity detection. Our diagnostic statistic is simple and is shown to track the transition from transience stationarity theoretically. We conduct extensive numerical experiments and compare our method against various existing approaches. Our proposed coupling-based stepsize scheme is observed to achieve superior performance across a diverse set of convex and non-convex problems. Moreover, our results demonstrate the robustness of our approach to a wide range of hyperparameters.
Due to the challenges in acquiring paired Text-3D data and the inherent irregularity of 3D data structures, combined representation learning of 3D point clouds and text remains unexplored. In this paper, we propose a novel Riemann-based Multi-scale Attention Reasoning Network (RMARN) for text-3D retrieval. Specifically, the extracted text and point cloud features are refined by their respective Adaptive Feature Refiner (AFR). Furthermore, we introduce the innovative Riemann Local Similarity (RLS) module and the Global Pooling Similarity (GPS) module. However, as 3D point cloud data and text data often possess complex geometric structures in high-dimensional space, the proposed RLS employs a novel Riemann Attention Mechanism to reflect the intrinsic geometric relationships of the data. Without explicitly defining the manifold, RMARN learns the manifold parameters to better represent the distances between text-point cloud samples. To address the challenges of lacking paired text-3D data, we have created the large-scale Text-3D Retrieval dataset T3DR-HIT, which comprises over 3,380 pairs of text and point cloud data. T3DR-HIT contains coarse-grained indoor 3D scenes and fine-grained Chinese artifact scenes, consisting of 1,380 and over 2,000 text-3D pairs, respectively. Experiments on our custom datasets demonstrate the superior performance of the proposed method.
High-level synthesis (HLS) is a widely used tool in designing Field Programmable Gate Array (FPGA). HLS enables FPGA design with software programming languages by compiling the source code into an FPGA circuit. The source code includes a program (called ``kernel'') and several pragmas that instruct hardware synthesis, such as parallelization, pipeline, etc. While it is relatively easy for software developers to design the program, it heavily relies on hardware knowledge to design the pragmas, posing a big challenge for software developers. Recently, different machine learning algorithms, such as GNNs, have been proposed to automate the pragma design via performance prediction. However, when applying the trained model on new kernels, the significant domain shift often leads to unsatisfactory performance. We propose a more domain-generalizable model structure: a two-level hierarchical Mixture of Experts (MoE), that can be flexibly adapted to any GNN model. Different expert networks can learn to deal with different regions in the representation space, and they can utilize similar patterns between the old kernels and new kernels. In the low-level MoE, we apply MoE on three natural granularities of a program: node, basic block, and graph. The high-level MoE learns to aggregate the three granularities for the final decision. To stably train the hierarchical MoE, we further propose a two-stage training method. Extensive experiments verify the effectiveness of the hierarchical MoE.
Unravelling Causal Genetic Biomarkers of Alzheimer’s Disease via Neuron to Gene-token Backtracking in Neural Architecture: A Groundbreaking Reverse-Gene-Finder Approach
PDF ↗Alzheimer’s Disease (AD) affects over 55 million people globally, yet the key genetic contributors remain poorly understood. Leveraging recent advancements in genomic foundation models, we present the innovative Reverse-Gene-Finder technology, a ground-breaking neuron-to-gene-token backtracking approach in a neural network architecture to elucidate the novel causal genetic biomarkers driving AD onset. Reverse-Gene-Finder comprises three key innovations. Firstly, we exploit the observation that genes with the highest probability of causing AD, defined as the most causal genes (MCGs), must have the highest probability of activating those neurons with the highest probability of causing AD, defined as the most causal neurons (MCNs). Secondly, we utilize a gene token representation at the input layer to allow each gene (known or novel to AD) to be represented as a discrete and unique entity in the input space. Lastly, in contrast to the existing neural network architectures, which track neuron activations from the input layer to the output layer in a feed-forward manner, we develop an innovative backtracking method to track backwards from the MCNs to the input layer, identifying the Most Causal Tokens (MCTs) and the corresponding MCGs. Reverse-Gene-Finder is highly interpretable, generalizable, and adaptable, providing a promising avenue for application in other disease scenarios.
Knowledge distillation (KD) aims to improve the performance of lightweight student networks under the guidance of pre-trained teachers. However, the large capacity gap between teachers and students limits the distillation gains. Previous methods addressing this problem have two weaknesses. First, most of them decrease the performance of pre-trained teachers, hindering students from achieving comparable performance. Second, these methods fail to dynamically adjust the transferred knowledge to be compatible with the representation ability of students, which is less effective in bridging the capacity gap. In this paper, we propose Adaptive Dual Guidance Knowledge Distillation (ADG-KD), which retains the guidance of the pre-trained teacher and uses the teacher's bidirectional optimization route guiding the student to alleviate the capacity gap problem. Specifically, ADG-KD introduces an initialized teacher, which has an identical structure to the pre-trained teacher and is optimized through the bidirectional supervision from both the pre-trained teacher and student. In this way, we construct the teacher's bidirectional optimization route to provide the students with an easy-to-hard and compatible knowledge sequence. ADG-KD trains the students under the proposed dual guidance approaches and automatically determines their importance weights, making the transferred knowledge better compatible with the representation ability of students. Extensive experiments on CIFAR-100, ImageNet, and MS-COCO demonstrate the effectiveness of our method.
Convolutional neural networks (CNNs) have been playing a dominant role in computer vision. However, the existing approaches of using local window modeling in popular CNNs lack flexibility and hinder their ability to capture long-range dependencies of objects in an image. To overcome these limitations, we propose a novel CNN architecture, termed Dynamic Clustering Convolutional Neural Network (DCCNeXt). The proposed DCCNeXt takes a unique approach by employing global clustering to group image patches with similar semantics into clusters that are then convolved using the shared convolution kernels. To address the high computational complexity of global clustering, the feature vectors from each patch's subspace are extracted for efficient clustering, which makes the proposed model widely compatible with the downstream vision tasks. The extensive experiments of image classification, object detection, instance segmentation, and semantic segmentation on the benchmark datasets demonstrate that the proposed DCCNeXt outperforms the mainstream Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), Vision Multi-layer Perceptrons (MLPs), Vision Graph Neural Networks (GNNs), and Vision Mambas. We anticipate that this study will provide a new perspective and a promising avenue for the design of convolutional neural networks.
We introduce a neural network conformal prediction method for time series that enhances adaptivity in non-stationary environments. Our approach acts as a neural controller designed to achieve desired target coverage, leveraging auxiliary multi-view data with neural network encoders in an end-to-end manner to further enhance adaptivity. Additionally, our model is designed to enhance the consistency of prediction intervals in different quantiles by integrating monotonicity constraints and leverages data from related tasks to boost few-shot learning performance. Using real-world datasets from epidemics, electric demand, weather, and others, we empirically demonstrate significant improvements in coverage and probabilistic accuracy, and find that our method is the only one that combines good calibration with consistency in prediction intervals.
Semi-Supervised Multi-View Multi-Label Learning with View-Specific Transformer and Enhanced Pseudo-Label
PDF ↗Multi-view multi-label learning has become a research focus for describing objects with rich expressions and annotations. However, real-world data often contains numerous unlabeled instances, due to the high cost and technical limitations of manual labeling. This crucial problem involves three main challenges: i) How to extract advanced semantics from available views? ii) How to build a refined classification framework with limited labeled space? iii) How to provide more high-quality supervisory information? To address these problems, we propose a Semi-Supervised Multi-View Multi-Label Learning Method with View-Specific Transformer and Enhanced Pseudo-Label named SMVTEP. Specifically, Generative Adversarial Networks are employed to extract informative shared and specific representations and their consistency and distinctiveness are ensured through the adversarial mechanism and information theory based contrastive learning. Then we build specific classifiers for each extracted feature and apply instance-level manifold constraints to reduce bias across classifiers. Moreover, we design a transformer-style fusion approach that simultaneously captures the imbalance of expressive power among views, mapping effects on specific labels, and label dependencies by incorporating confidence scores and category semantics into the self-attention mechanism. Furthermore, after using Mixup for data augmentation, category-enhanced pseudo-labels are leveraged to improve the reliability of additional annotations by aligning the label distribution of unlabeled samples with the true distribution. Finally, extensive experimental results validate the effectiveness of SMVTEP against state-of-the-art methods.
Recently, prompt tuning methods for pre-trained models have demonstrated promising performance in Class Incremental Learning (CIL). These methods typically involve learning task-specific prompts and predicting the task ID to select the appropriate prompts for inference. However, inaccurate task ID predictions can cause severe inconsistencies between the prompts used during training and inference, leading to knowledge forgetting and performance degradation. Additionally, existing prompt tuning methods rely solely on the pre-trained model to predict task IDs, without fully leveraging the knowledge embedded in the learned prompt parameters, resulting in inferior prediction performance. To address these issues, we propose a novel Cyclic Prompt Aggregation (CAPrompt) method that eliminates the dependency on task ID prediction by cyclically aggregating the knowledge from different prompts. Specifically, rather than predicting task IDs, we introduce an innovative prompt aggregation strategy during both training and inference to overcome prompt inconsistency by utilizing a weighted sum of different prompts. Thorough theoretical analysis demonstrates that under concave conditions, the aggregated prompt achieves lower error compared to selecting a single task-specific prompt. Consequently, we incorporate a concave constraint and a linear constraint to guide prompt learning, ensuring compliance with the concave condition requirement. Furthermore, to fully exploit the prompts and achieve more accurate prompt weights, we develop a cyclic weight prediction strategy. This strategy begins with equal weights for each task and automatically adjusts them to more appropriate values in a cyclical manner. Experiments on various datasets demonstrate that our proposed CAPrompt outperforms state-of-the-art methods by 2%-3%.
Unmanned Aerial Vehicle Object Detection (UAVOD) presents unique challenges due to varying altitudes, dynamic backgrounds, and the small size of objects. Traditional detection methods often struggle with these challenges, as they typically rely on visual feature only and fail to extract the semantic relations between the objects. To address these limitations, we propose a novel approach named Self-Prompting Analogical Reasoning (SPAR). Our method utilizes the vision-language model (CLIP) to generate context-aware prompts based on image feature, providing rich semantic information that guides analogical reasoning. SPAR includes two main modules: self-prompting and analogical reasoning. Self-prompting module based on learnable description and CLIP-text encoder generates context-aware prompt by combining specific image feature; then an objectness prompt score map is produced by computing the similarity between pixel-level features and context-aware prompt. With this score map, multi-scale image features are enhanced and pixel-level features are chosen for graph construction. While for analogical reasoning module, graph nodes consists of category-level prompt nodes and pixel-level image feature nodes. Analogical inference is based graph convolution. Under the guidance of category-level nodes, different-scale object features have been enhanced, which helps achieve more accurate detection of challenging objects. Extensive experiments illustrate that SPAR outperforms traditional methods, offering a more robust and accurate solution for UAVOD.
Convolution-based unlearnable examples (UEs) employ class-wise multiplicative convolutional noise to training samples, severely compromising model performance. This fire-new type of UEs have successfully countered all defense mechanisms against UEs. The failure of such defenses can be attributed to the absence of norm constraints on convolutional noise, leading to severe blurring of image features. To address this, we first design an Edge Pixel-based Detector (EPD) to identify convolution-based UEs. Upon detection of them, we propose the first defense scheme against convolution-based UEs, COrrupting these samples via random matrix multiplication by employing bilinear INterpolation (COIN) such that disrupting the distribution of class-wise multiplicative noise. To evaluate the generalization of our proposed COIN, we newly design two convolution-based UEs called VUDA and HUDA to expand the scope of convolution-based UEs. Extensive experiments demonstrate the effectiveness of detection scheme EPD and that our defense COIN outperforms 11 state-of-the-art (SOTA) defenses, achieving a significant improvement on the CIFAR and ImageNet datasets.
The transformer architecture has catalyzed revolutionary advances in language modeling. However, recent architectural recipes, such as state-space models, have bridged the performance gap. Motivated by this, we examine the benefits of Convolution-Augmented Transformer (CAT) for recall, copying, and length generalization tasks. CAT incorporates convolutional filters in the K/Q/V embeddings of an attention layer. Through CAT, we show that the locality of the convolution synergizes with the global view of the attention. Unlike comparable architectures, such as Mamba or transformer, CAT can provably solve the associative recall (AR) and copying tasks using a single layer while also enjoying guaranteed length generalization. We also establish computational tradeoffs between convolution and attention by characterizing how convolution can mitigate the need for full attention by summarizing the context window and creating salient summary tokens to attend. Evaluations on real datasets corroborate our findings and demonstrate that CAT and its variations indeed enhance the language modeling performance.
Hypergraphs provide a flexible framework for modeling high-order (complex) interactions among multiple entities, extending beyond traditional pairwise correlations in graph structures. However, deep hypergraph neural networks (HGNNs) often face the challenge of oversmoothing with increasing depth, similar to issues in graph neural networks (GNNs). While oversmoothing in GNNs has been extensively studied, its implications in relation to hypergraphs are less explored. This paper addresses this gap by first theoretically exploring the reasons behind oversmoothing in deep HGNNs. Our novel insights suggest that a spectral-based hypergraph convolution, equipped with both low-pass and high-pass filters, can potentially mitigate these effects. Motivated by these findings, we introduce FrameHGNN, a framework that utilizes framelet-based hypergraph convolutions integrating tight framelet transforms with both low-pass and high-pass components, as well as the commonly used strategies in designing deep GNN architecture: initial residual and identity mappings. The experiment results on diverse benchmark datasets demonstrate that FrameHGNN outperforms several state-of-the-art models, effectively reducing oversmoothing while improving predictive accuracy. Our contributions not only advance the theoretical understanding of deep hypergraph learning but also provide a practical spectral-based approach for HGNNs, emphasizing the design of multifrequency channels.
Hypergraph neural networks (HNNs) have shown promise in handling tasks characterized by high-order correlations, achieving notable success across various applications. However, there has been limited focus on heterophilic hypergraph learning (HHL), in contrast to the increasing attention given to graph neural networks designed for graphs exhibiting heterophily. This paper aims to pave the way for HHL by addressing key gaps from multiple perspectives: measurement, dataset diversity, and baseline model development. First, we introduce metrics to quantify heterophily in hypergraphs, providing a numerical basis for assessing the homophily/heterophily ratio. Second, we develop diverse benchmark datasets across various real-world scenarios, facilitating comprehensive evaluations of existing HNNs and advancing research in HHL. Additionally, as a novel baseline model, we propose HyperUFG, a framelet-based HNN integrating both low-pass and high-pass filters. Extensive experiments conducted on synthetic and benchmark datasets highlight the challenges current HNNs face with heterophilic hypergraphs, while showcasing that HyperUFG performs competitively and often outperforms many existing models in such scenarios. Overall, our study underscores the urgent need for further exploration and development in this emerging field, with the potential to inspire and guide future research in HHL.
The rapid advancement of multi-agent reinforcement learning (MARL) has given rise to diverse training paradigms to learn the policies of each agent in the multi-agent system. The paradigms of decentralized training and execution (DTDE) and centralized training with decentralized execution (CTDE) have been proposed and widely applied. However, as the number of agents increases, the inherent limitations of these frameworks significantly degrade the performance metrics, such as win rate, total reward, etc. To reduce the influence of the increasing number of agents on the performance metrics, we propose a novel training paradigm of grouped training decentralized execution (GTDE). This framework eliminates the need for a centralized module and relies solely on local information, effectively meeting the training requirements of large-scale multi-agent systems. Specifically, we first introduce an adaptive grouping module, which divides each agent into different groups based on their observation history. To implement end-to-end training, GTDE uses Gumbel-Sigmoid for efficient point-to-point sampling on the grouping distribution while ensuring gradient backpropagation. To adapt to the uncertainty in the number of members in a group, two methods are used to implement a group information aggregation module that merges member information within the group. Empirical results show that in a cooperative environment with 495 agents, GTDE increased the total reward by an average of 382% compared to the baseline. In a competitive environment with 64 agents, GTDE achieved a 100% win rate against the baseline.
Few-Shot Class-Incremental Learning (FSCIL) aims to continuously learn new classes from a limited set of training samples without forgetting knowledge of previously learned classes. Conventional FSCIL methods typically build a robust feature extractor during the base training session with abundant training samples and subsequently freeze this extractor, only fine-tuning the classifier in subsequent incremental phases. However, current strategies primarily focus on preventing catastrophic forgetting, considering only the relationship between novel and base classes, without paying attention to the specific decision spaces of each class. To address this challenge, we propose a plug-and-play Adaptive Decision Boundary Strategy (ADBS), which is compatible with most FSCIL methods. Specifically, we assign a specific decision boundary to each class and adaptively adjust these boundaries during training to optimally refine the decision spaces for the classes in each session. Furthermore, to amplify the distinctiveness between classes, we employ a novel inter-class constraint loss that optimizes the decision boundaries and prototypes for each class. Extensive experiments on three benchmarks, namely CIFAR100, miniImageNet, and CUB200, demonstrate that incorporating our ADBS method with existing FSCIL techniques significantly improves performance, achieving overall state-of-the-art results.