论文检索

Pingting Hao, Kunpeng Liu, Wanfu Gao

Multi-view multi-label feature selection aims to select informative features where the data are collected from multiple sources with multiple interdependent class labels. For fully exploiting multi-view information, most prior works mainly focus on the common part in the ideal circumstance. However, the inconsistent part hidden in each view, including noises and specific elements, may affect the quality of mapping between labels and feature representations. Meanwhile, ignoring the specific part might lead to a suboptimal result, as each label is supposed to possess specific characteristics of its own. To deal with the double problems in multi-view multi-label feature selection, we propose a unified loss function which is a totally splitting structure for observed labels as hybrid labels that is, common labels, view-to-all specific labels and noisy labels, and the view-to-all specific labels further splits into several specific labels of each view. The proposed method simultaneously considers the consistency and complementarity of different views. Through exploring the feature weights of hybrid labels, the mapping relationships between labels and features can be established sequentially based on their attributes. Additionally, the interrelatedness among hybrid labels is also investigated and injected into the function. Specific to the specific labels of each view, we construct the novel regularization paradigm incorporating logic operations. Finally, the convergence of the result is proved after applying the multiplicative update rules. Experiments on six datasets demonstrate the effectiveness and superiority of our method compared with the state-of-the-art methods.

Guang-Yuan Hao, Hengguan Huang, Haotian Wang, Jie Gao, Hao Wang

Active learning (AL) aims to improve model performance within a fixed labeling budget by choosing the most informative data points to label. Existing AL focuses on the single-domain setting, where all data come from the same domain (e.g., the same dataset). However, many real-world tasks often involve multiple domains. For example, in visual recognition, it is often desirable to train an image classifier that works across different environments (e.g., different backgrounds), where images from each environment constitute one domain. Such a multi-domain AL setting is challenging for prior methods because they (1) ignore the similarity among different domains when assigning labeling budget and (2) fail to handle distribution shift of data across different domains. In this paper, we propose the first general method, dubbed composite active learning (CAL), for multi-domain AL. Our approach explicitly considers the domain-level and instance-level information in the problem; CAL first assigns domain-level budgets according to domain-level importance, which is estimated by optimizing an upper error bound that we develop; with the domain-level budgets, CAL then leverages a certain instance-level query strategy to select samples to label from each domain. Our theoretical analysis shows that our method achieves a better error bound compared to current AL methods. Our empirical results demonstrate that our approach significantly outperforms the state-of-the-art AL methods on both synthetic and real-world multi-domain datasets. Code is available at https://github.com/Wang-ML-Lab/multi-domain-active-learning.

Teemu Hankala, Miika Hannula, Juha Kontinen, Jonni Virtema

The training problem of neural networks (NNs) is known to be ER-complete with respect to ReLU and linear activation functions. We show that the training problem for NNs equipped with arbitrary activation functions is polynomial-time bireducible to the existential theory of the reals extended with the corresponding activation functions. For effectively continuous activation functions (e.g., the sigmoid function), we obtain an inclusion to low levels of the arithmetical hierarchy. Consequently, the sigmoid activation function leads to the existential theory of the reals with the exponential function, and hence the decidability of training NNs using the sigmoid activation function is equivalent to the decidability of the existential theory of the reals with the exponential function, a long-standing open problem. In contrast, we obtain that the training problem is undecidable if sinusoidal activation functions are considered.

Qi Han, Li Zhu, Fei Guo

The multi-armed bandit(MAB) is a classical sequential decision problem. Most work requires assumptions about the reward distribution (e.g., bounded), while practitioners may have difficulty obtaining information about these distributions to design models for their problems, especially in non-stationary MAB problems. This paper aims to design a multi-armed bandit algorithm that can be implemented without using information about the reward distribution while still achieving substantial regret upper bounds. To this end, we propose a novel algorithm alternating between greedy rule and forced exploration. Our method can be applied to Gaussian, Bernoulli and other subgaussian distributions, and its implementation does not require additional information. We employ a unified analysis method for different forced exploration strategies and provide problem-dependent regret upper bounds for stationary and piecewise-stationary settings. Furthermore, we compare our algorithm with popular bandit algorithms on different reward distributions.

Seokhyeon Ha, Sunbeom Jeong, Jungwoo Lee

Fine-tuning pre-trained neural network models has become a widely adopted approach across various domains. However, it can lead to the distortion of pre-trained feature extractors that already possess strong generalization capabilities. Mitigating feature distortion during adaptation to new target domains is crucial. Recent studies have shown promising results in handling feature distortion by aligning the head layer on in-distribution datasets before performing fine-tuning. Nonetheless, a significant limitation arises from the treatment of batch normalization layers during fine-tuning, leading to suboptimal performance. In this paper, we propose Domain-Aware Fine-Tuning (DAFT), a novel approach that incorporates batch normalization conversion and the integration of linear probing and fine-tuning. Our batch normalization conversion method effectively mitigates feature distortion by reducing modifications to the neural network during fine-tuning. Additionally, we introduce the integration of linear probing and fine-tuning to optimize the head layer with gradual adaptation of the feature extractor. By leveraging batch normalization layers and integrating linear probing and fine-tuning, our DAFT significantly mitigates feature distortion and achieves improved model performance on both in-distribution and out-of-distribution datasets. Extensive experiments demonstrate that our method outperforms other baseline methods, demonstrating its effectiveness in not only improving performance but also mitigating feature distortion.

Dhawal Gupta, Scott M. Jordan, Shreyas Chaudhari, Bo Liu, Philip S. Thomas, Bruno Castro da Silva

In this paper, we introduce a fresh perspective on the challenges of credit assignment and policy evaluation. First, we delve into the nuances of eligibility traces and explore instances where their updates may result in unexpected credit assignment to preceding states. From this investigation emerges the concept of a novel value function, which we refer to as the ????????????? ????? ????????. Unlike traditional state value functions, bidirectional value functions account for both future expected returns (rewards anticipated from the current state onward) and past expected returns (cumulative rewards from the episode's start to the present). We derive principled update equations to learn this value function and, through experimentation, demonstrate its efficacy in enhancing the process of policy evaluation. In particular, our results indicate that the proposed learning approach can, in certain challenging contexts, perform policy evaluation more rapidly than TD(λ)–a method that learns forward value functions, v^π, ????????. Overall, our findings present a new perspective on eligibility traces and potential advantages associated with the novel value function it inspires, especially for policy evaluation.

Yufei Guo, Yuanpei Chen, Xiaode Liu, Weihang Peng, Yuhan Zhang, Xuhui Huang, Zhe Ma

The Spiking Neural Network (SNN), as one of the biologically inspired neural network infrastructures, has drawn increasing attention recently. It adopts binary spike activations to transmit information, thus the multiplications of activations and weights can be substituted by additions, which brings high energy efficiency. However, in the paper, we theoretically and experimentally prove that the binary spike activation map cannot carry enough information, thus causing information loss and resulting in accuracy decreasing. To handle the problem, we propose a ternary spike neuron to transmit information. The ternary spike neuron can also enjoy the event-driven and multiplication-free operation advantages of the binary spike neuron but will boost the information capacity. Furthermore, we also embed a trainable factor in the ternary spike neuron to learn the suitable spike amplitude, thus our SNN will adopt different spike amplitudes along layers, which can better suit the phenomenon that the membrane potential distributions are different along layers. To retain the efficiency of the vanilla ternary spike, the trainable ternary spike SNN will be converted to a standard one again via a re-parameterization technique in the inference. Extensive experiments with several popular network structures over static and dynamic datasets show that the ternary spike can consistently outperform state-of-the-art methods. Our code is open-sourced at https://github.com/yfguo91/Ternary-Spike.

Xianjie Guo, Kui Yu, Lin Liu, Jiuyong Li

As an emerging research direction, federated causal structure learning (CSL) aims at learning causal relationships from decentralized data across multiple clients while preserving data privacy. Existing federated CSL algorithms suffer from scalability and accuracy issues, since they require computationally expensive CSL algorithms to be executed at each client. Furthermore, in real-world scenarios, the number of samples held by each client varies significantly, and existing methods still assign equal weights to the learned structural information from each client, which severely harms the learning accuracy of those methods. To address these two limitations, we propose FedCSL, a scalable and accurate method for federated CSL. Specifically, FedCSL consists of two novel strategies: (1) a federated local-to-global learning strategy that enables FedCSL to scale to high-dimensional data for tackling the scalability issue, and (2) a novel weighted aggregation strategy that does not rely on any complex encryption techniques while preserving data privacy for tackling the accuracy issue. Extensive experiments on benchmark datasets, high-dimensional synthetic datasets and a real-world dataset verify the efficacy of the proposed FedCSL method. The source code is available at https://github.com/Xianjie-Guo/FedCSL.

Anchun Gui, Jinqiang Ye, Han Xiao

It has become a popular paradigm to transfer the knowledge of large-scale pre-trained models to various downstream tasks via fine-tuning the entire model parameters. However, with the growth of model scale and the rising number of downstream tasks, this paradigm inevitably meets the challenges in terms of computation consumption and memory footprint issues. Recently, Parameter-Efficient Fine-Tuning (PEFT) (e.g., Adapter, LoRA, BitFit) shows a promising paradigm to alleviate these concerns by updating only a portion of parameters. Despite these PEFTs having demonstrated satisfactory performance in natural language processing, it remains under-explored for the question: whether these techniques could be transferred to graph-based tasks with Graph Transformer Networks (GTNs)? Therefore, in this paper, we fill this gap by providing extensive benchmarks with traditional PEFTs on a range of graph-based downstream tasks. Our empirical study shows that it is sub-optimal to directly transfer existing PEFTs to graph-based tasks due to the issue of feature distribution shift. To address this issue, we propose a novel structure-aware PEFT approach, named G-Adapter, which leverages graph convolution operation to introduce graph structure information (e.g., graph adjacency matrix) as an inductive bias to guide the updating process. Further, we propose Bregman proximal point optimization to alleviate feature distribution shift by preventing the model from aggressive update. Extensive experiments demonstrate that G-Adapter obtains state-of-the-art performance compared to counterparts on nine graph benchmark datasets based on diverse pre-trained GTNs, and delivers tremendous memory footprint efficiency compared to the conventional paradigm.

Jianyang Gu, Kai Wang, Wei Jiang, Yang You

Replay-based methods have proved their effectiveness on online continual learning by rehearsing past samples from an auxiliary memory. With many efforts made on improving training schemes based on the memory, however, the information carried by each sample in the memory remains under-investigated. Under circumstances with restricted storage space, the informativeness of the memory becomes critical for effective replay. Although some works design specific strategies to select representative samples, by only employing a small number of original images, the storage space is still not well utilized. To this end, we propose to Summarize the knowledge from the Stream Data (SSD) into more informative samples by distilling the training characteristics of real images. Through maintaining the consistency of training gradients and relationship to the past tasks, the summarized samples are more representative for the stream data compared to the original images. Extensive experiments are conducted on multiple online continual learning benchmarks to support that the proposed SSD method significantly enhances the replay effects. We demonstrate that with limited extra computational overhead, SSD provides more than 3% accuracy boost for sequential CIFAR-100 under extremely restricted memory buffer. Code in https://github.com/vimar-gu/SSD.

Andreas Grivas, Antonio Vergari, Adam Lopez

Sigmoid output layers are widely used in multi-label classification (MLC) tasks, in which multiple labels can be assigned to any input. In many practical MLC tasks, the number of possible labels is in the thousands, often exceeding the number of input features and resulting in a low-rank output layer. In multi-class classification, it is known that such a low-rank output layer is a bottleneck that can result in unargmaxable classes: classes which cannot be predicted for any input. In this paper, we show that for MLC tasks, the analogous sigmoid bottleneck results in exponentially many unargmaxable label combinations. We explain how to detect these unargmaxable outputs and demonstrate their presence in three widely used MLC datasets. We then show that they can be prevented in practice by introducing a Discrete Fourier Transform (DFT) output layer, which guarantees that all sparse label combinations with up to k active labels are argmaxable. Our DFT layer trains faster and is more parameter efficient, matching the F1@k score of a sigmoid layer while using up to 50% fewer trainable parameters. Our code is publicly available at https://github.com/andreasgrv/sigmoid-bottleneck.

Kshitij Goyal, Sebastijan Dumancic, Hendrik Blockeel

As machine learning models, specifically neural networks, are becoming increasingly popular, there are concerns regarding their trustworthiness, specially in safety-critical applications, e.g. actions of an autonomous vehicle must be safe. There are approaches that can train neural networks where such domain requirements are enforced as constraints, but they either cannot guarantee that the constraint will be satisfied by all possible predictions (even on unseen data) or they are limited in the type of constraints that can be enforced. In this paper, we present an approach to train neural networks which can enforce a wide variety of constraints and guarantee that the constraint is satisfied by all possible predictions. The approach builds on earlier work where learning linear models is formulated as a constraint satisfaction problem (CSP). To make this idea applicable to neural networks, two crucial new elements are added: constraint propagation over the network layers, and weight updates based on a mix of gradient descent and CSP solving. Evaluation on various machine learning tasks demonstrates that our approach is flexible enough to enforce a wide variety of domain constraints and is able to guarantee them in neural networks.

Ruihao Gong, Yang Yong, Zining Wang, Jinyang Guo, Xiuying Wei, Yuqing Ma, Xianglong Liu

Neural network sparsity has attracted many research interests due to its similarity to biological schemes and high energy efficiency. However, existing methods depend on long-time training or fine-tuning, which prevents large-scale applications. Recently, some works focusing on post-training sparsity (PTS) have emerged. They get rid of the high training cost but usually suffer from distinct accuracy degradation due to neglect of the reasonable sparsity rate at each layer. Previous methods for finding sparsity rates mainly focus on the training-aware scenario, which usually fails to converge stably under the PTS setting with limited data and much less training cost. In this paper, we propose a fast and controllable post-training sparsity (FCPTS) framework. By incorporating a differentiable bridge function and a controllable optimization objective, our method allows for rapid and accurate sparsity allocation learning in minutes, with the added assurance of convergence to a predetermined global sparsity rate. Equipped with these techniques, we can surpass the state-of-the-art methods by a large margin, e.g., over 30\% improvement for ResNet-50 on ImageNet under the sparsity rate of 80\%. Our plug-and-play code and supplementary materials are open-sourced at https://github.com/ModelTC/FCPTS.

Chengyue Gong, Xiaocong Du, Bhargav Bhushanam, Lemeng Wu, Xingchao Liu, Dhruv Choudhary, Arun Kejariwal, Qiang Liu

Very deep neural networks lead to significantly better performance on various real tasks. However, it usually causes slow inference and is hard to be deployed on real-world devices. How to reduce the number of layers to save memory and to accelerate the inference is an eye-catching topic. In this work, we introduce an intermediate objective, a continuous-time network, before distilling deep networks into shallow networks. First, we distill a given deep network into a continuous-time neural flow model, which can be discretized with an ODE solver and the inference requires passing through the network multiple times. By forcing the flow transport trajectory to be straight lines, we find that it is easier to compress the infinite step model into a one-step neural flow model, which only requires passing through the flow model once. Secondly, we refine the one-step flow model together with the final head layer with knowledge distillation and finally, we can replace the given deep network with this one-step flow network. Empirically, we demonstrate that our method outperforms direct distillation and other baselines on different model architectures (e.g. ResNet, ViT) on image classification and semantic segmentation tasks. We also manifest that our distilled model naturally serves as an early-exit dynamic inference model.

Joseph Giovanelli, Alexander Tornede, Tanja Tornede, Marius Lindauer

Hyperparameter optimization (HPO) is important to leverage the full potential of machine learning (ML). In practice, users are often interested in multi-objective (MO) problems, i.e., optimizing potentially conflicting objectives, like accuracy and energy consumption. To tackle this, the vast majority of MO-ML algorithms return a Pareto front of non-dominated machine learning models to the user. Optimizing the hyperparameters of such algorithms is non-trivial as evaluating a hyperparameter configuration entails evaluating the quality of the resulting Pareto front. In literature, there are known indicators that assess the quality of a Pareto front (e.g., hypervolume, R2) by quantifying different properties (e.g., volume, proximity to a reference point). However, choosing the indicator that leads to the desired Pareto front might be a hard task for a user. In this paper, we propose a human-centered interactive HPO approach tailored towards multi-objective ML leveraging preference learning to extract desiderata from users that guide the optimization. Instead of relying on the user guessing the most suitable indicator for their needs, our approach automatically learns an appropriate indicator. Concretely, we leverage pairwise comparisons of distinct Pareto fronts to learn such an appropriate quality indicator. Then, we optimize the hyperparameters of the underlying MO-ML algorithm towards this learned indicator using a state-of-the-art HPO approach. In an experimental study targeting the environmental impact of ML, we demonstrate that our approach leads to substantially better Pareto fronts compared to optimizing based on a wrong indicator pre-selected by the user, and performs comparable in the case of an advanced user knowing which indicator to pick.

Anne-Marie George, Christos Dimitrakakis

We formulate the problem of eliciting agents' preferences with the goal of finding a Kemeny ranking as a Dueling Bandits problem. Here the bandits' arms correspond to alternatives that need to be ranked and the feedback corresponds to a pairwise comparison between alternatives by a randomly sampled agent. We consider both sampling with and without replacement, i.e., the possibility to ask the same agent about some comparison multiple times or not. We find approximation bounds for Kemeny rankings dependant on confidence intervals over estimated winning probabilities of arms. Based on these we state algorithms to find Probably Approximately Correct (PAC) solutions and elaborate on their sample complexity for sampling with or without replacement. Furthermore, if all agents' preferences are strict rankings over the alternatives, we provide means to prune confidence intervals and thereby guide a more efficient elicitation. We formulate several adaptive sampling methods that use look-aheads to estimate how much confidence intervals (and thus approximation guarantees) might be tightened. All described methods are compared on synthetic data.

Shengyi Gao, Zhe Chen, Guo Chen, Wenhai Wang, Tong Lu

Audio-visual segmentation (AVS) aims to locate and segment the sounding objects in a given video, which demands audio-driven pixel-level scene understanding. The existing methods cannot fully process the fine-grained correlations between audio and visual cues across various situations dynamically. They also face challenges in adapting to complex scenarios, such as evolving audio, the coexistence of multiple objects, and more. In this paper, we propose AVSegFormer, a novel framework for AVS that leverages the transformer architecture. Specifically, It comprises a dense audio-visual mixer, which can dynamically adjust interested visual features, and a sparse audio-visual decoder, which implicitly separates audio sources and automatically matches optimal visual features. Combining both components provides a more robust bidirectional conditional multi-modal representation, improving the segmentation performance in different scenarios. Extensive experiments demonstrate that AVSegFormer achieves state-of-the-art results on the AVS benchmark. The code is available at https://github.com/vvvb-github/AVSegFormer.

Hang Gao, Chengyu Yao, Jiangmeng Li, Lingyu Si, Yifan Jin, Fengge Wu, Changwen Zheng, Huaping Liu

Graph Neural Networks (GNNs) demonstrate their significance by effectively modeling complex interrelationships within graph-structured data. To enhance the credibility and robustness of GNNs, it becomes exceptionally crucial to bolster their ability to capture causal relationships. However, despite recent advancements that have indeed strengthened GNNs from a causal learning perspective, conducting an in-depth analysis specifically targeting the causal modeling prowess of GNNs remains an unresolved issue. In order to comprehensively analyze various GNN models from a causal learning perspective, we constructed an artificially synthesized dataset with known and controllable causal relationships between data and labels. The rationality of the generated data is further ensured through theoretical foundations. Drawing insights from analyses conducted using our dataset, we introduce a lightweight and highly adaptable GNN module designed to strengthen GNNs' causal learning capabilities across a diverse range of tasks. Through a series of experiments conducted on both synthetic datasets and other real-world datasets, we empirically validate the effectiveness of the proposed module. The codes are available at https://github.com/yaoyao-yaoyao-cell/CRCG.

Ge Gao, Xi Yang, Min Chi

Reinforcement learning (RL) is broadly employed in human-involved systems to enhance human outcomes. Off-policy evaluation (OPE) has been pivotal for RL in those realms since online policy learning and evaluation can be high-stake. Intelligent tutoring has raised tremendous attentions as highly challenging when applying OPE to human-involved systems, due to that students' subgroups can favor different pedagogical policies and the costly procedure that policies have to be induced fully offline and then directly deployed to the upcoming semester. In this work, we formulate on-demand pedagogical policy selection (ODPS) to tackle the challenges for OPE in intelligent tutoring. We propose a pipeline, EduPlanner, as a concrete solution for ODPS. Our pipeline results in an theoretically unbiased estimator, and enables efficient and customized policy selection by identifying subgroups over both historical data and on-arrival initial logs. We evaluate our approach on the Probability ITS that has been used in real classrooms for over eight years. Our study shows significant improvement on learning outcomes of students with EduPlanner, especially for the ones associated with low-performing subgroups.

Chen-Xiao Gao, Chenyang Wu, Mingjun Cao, Rui Kong, Zongzhang Zhang, Yang Yu

Decision Transformer (DT), which employs expressive sequence modeling techniques to perform action generation, has emerged as a promising approach to offline policy optimization. However, DT generates actions conditioned on a desired future return, which is known to bear some weaknesses such as the susceptibility to environmental stochasticity. To overcome DT's weaknesses, we propose to empower DT with dynamic programming. Our method comprises three steps. First, we employ in-sample value iteration to obtain approximated value functions, which involves dynamic programming over the MDP structure. Second, we evaluate action quality in context with estimated advantages. We introduce two types of advantage estimators, IAE and GAE, which are suitable for different tasks. Third, we train an Advantage-Conditioned Transformer (ACT) to generate actions conditioned on the estimated advantages. Finally, during testing, ACT generates actions conditioned on a desired advantage. Our evaluation results validate that, by leveraging the power of dynamic programming, ACT demonstrates effective trajectory stitching and robust action generation in spite of the environmental stochasticity, outperforming baseline methods across various benchmarks. Additionally, we conduct an in-depth analysis of ACT's various design choices through ablation studies. Our code is available at https://github.com/LAMDA-RL/ACT.