Search

Meta-learning enables learning systems to adapt quickly to new tasks, similar to humans. To emulate this human-like rapid learning and enhance alignment and discrimination abilities, we propose ConML, a universal meta-learning framework that can be applied to various meta-learning algorithms without relying on specific model architectures nor target models. The core of ConML is task-level contrastive learning, which extends contrastive learning from the representation space in unsupervised learning to the model space in meta-learning. By leveraging task identity as an additional supervision signal during meta-training, we contrast the outputs of the meta-learner in the model space, minimizing inner-task distance (between models trained on different subsets of the same task) and maximizing inter-task distance (between models from different tasks). We demonstrate that ConML integrates seamlessly with optimization-based, metric-based, and amortization-based meta-learning algorithms, as well as in-context learning, resulting in performance improvements across diverse few-shot learning tasks.

In-context learning (ICL) enables large language models (LLMs) to generalize to new tasks by incorporating a few in-context examples (ICEs) directly in the input, without updating parameters. However, the effectiveness of ICL heavily relies on the selection of ICEs, and conventional text-based embedding methods are often inadequate for tasks that require multi-step reasoning, such as mathematical and logical problem solving. This is due to the bias introduced by shallow semantic similarities that fail to capture the deeper reasoning structures required for these tasks. We present GraphIC, a novel approach that leverages graph-based representations of reasoning processes, coupled with Bayesian Networks (BNs) to select ICEs. Graph structures inherently filter out shallow semantics while preserving the core reasoning structure. Importantly, BNs capture the dependency of a node's attributes on its parent nodes, closely mirroring the hierarchical nature of human cognition-where each thought is shaped by preceding ones. This makes BNs particularly well-suited for multi-step reasoning tasks, aligning the process more closely with human-like reasoning. Extensive experiments across three types of reasoning tasks (mathematical reasoning, code generation, and logical reasoning) demonstrate that GraphIC outperforms both training-free and training-based models in selecting ICEs, excelling in terms of both effectiveness and efficiency. We show that GraphIC enhances ICL's performance and interoperability, significantly advancing ICE selection for multi-step reasoning tasks.

The local and global well-posedness for the one dimensional fourth-order nonlinear Schr\"odinger equation are established in the modulation space $M^{s}_{2,q}$ for $s\geq \frac12$ and $2\leq q <\infty$. The local result is based on the $U^p-V^p$ spaces and crucial bilinear estimates. The key ingredient to obtain the global well-posedness is that we achieve a-priori estimates of the solution in modulation spaces by utilizing the power series expansion of the perturbation determinant introduced by Killip-Visan-Zhang for completely integrable PDEs., Comment: 41 pages

The request for fast response and safe operation after natural and man-made disasters in urban environments has spurred the development of robotic systems designed to assist in search and rescue operations within complex rubble sites. Traditional Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) face significant limitations in such confined and obstructed environments. This paper introduces a novel vine robot designed to navigate dense rubble, drawing inspiration from natural growth mechanisms found in plants. Unlike conventional robots, vine robots are soft robots that can grow by everting their material, allowing them to navigate through narrow spaces and obstacles. The prototype presented in this study incorporates pneumatic muscles for steering and oscillation, an equation-based robot length control plus feedback pressure regulating system for extending and retracting the robot body. We conducted a series of controlled experiments in an artificial rubble testbed to assess the robot performance under varying environmental conditions and robot parameters, including volume ratio, environmental weight, oscillation, and steering. The results show that the vine robot can achieve significant penetration depths in cluttered environments with mixed obstacle sizes and weights, and can maintain repeated trajectories, demonstrating potential for mapping and navigating complex underground paths. Our findings highlight the suitability of the vine robot for urban search and rescue missions, with further research planned to enhance its robustness and deployability in real-world scenarios., Comment: Upon further review, this research was conducted as part of a short-term project, and in hindsight, it does not offer the level of depth and exhaustiveness necessary for a complete study. It would be in the best interest of the academic community to withdraw the paper at this time.

In federated learning (FL), accommodating clients' varied computational capacities poses a challenge, often limiting the participation of those with constrained resources in global model training. To address this issue, the concept of model heterogeneity through submodel extraction has emerged, offering a tailored solution that aligns the model's complexity with each client's computational capacity. In this work, we propose Federated Importance-Aware Submodel Extraction (FIARSE), a novel approach that dynamically adjusts submodels based on the importance of model parameters, thereby overcoming the limitations of previous static and dynamic submodel extraction methods. Compared to existing works, the proposed method offers a theoretical foundation for the submodel extraction and eliminates the need for additional information beyond the model parameters themselves to determine parameter importance, significantly reducing the overhead on clients. Extensive experiments are conducted on various datasets to showcase superior performance of the proposed FIARSE.

In recommendation systems, new items are continuously introduced, initially lacking interaction records but gradually accumulating them over time. Accurately predicting the click-through rate (CTR) for these items is crucial for enhancing both revenue and user experience. While existing methods focus on enhancing item ID embeddings for new items within general CTR models, they tend to adopt a global feature interaction approach, often overshadowing new items with sparse data by those with abundant interactions. Addressing this, our work introduces EmerG, a novel approach that warms up cold-start CTR prediction by learning item-specific feature interaction patterns. EmerG utilizes hypernetworks to generate an item-specific feature graph based on item characteristics, which is then processed by a Graph Neural Network (GNN). This GNN is specially tailored to provably capture feature interactions at any order through a customized message passing mechanism. We further design a meta learning strategy that optimizes parameters of hypernetworks and GNN across various item CTR prediction tasks, while only adjusting a minimal set of item-specific parameters within each task. This strategy effectively reduces the risk of overfitting when dealing with limited data. Extensive experiments on benchmark datasets validate that EmerG consistently performs the best given no, a few and sufficient instances of new items., Comment: KDD 2024

The scaling law, which involves the brute-force expansion of training datasets and learnable parameters, has become a prevalent strategy for developing more robust learning models. However, due to bottlenecks in data, computation, and trust, the sustainability of the scaling law is a serious concern for the future of deep learning. In this paper, we address this issue by developing next-generation models in a parsimonious manner (i.e., achieving greater potential with simpler models). The key is to drive models using domain-specific knowledge, such as symbols, logic, and formulas, instead of relying on the scaling law. This approach allows us to build a framework that uses this knowledge as "building blocks" to achieve parsimony in model design, training, and interpretation. Empirical results show that our methods surpass those that typically follow the scaling law. We also demonstrate the application of our framework in AI for science, specifically in the problem of drug-drug interaction prediction. We hope our research can foster more diverse technical roadmaps in the era of foundation models.

Synthesizing electronic health records (EHR) data has become a preferred strategy to address data scarcity, improve data quality, and model fairness in healthcare. However, existing approaches for EHR data generation predominantly rely on state-of-the-art generative techniques like generative adversarial networks, variational autoencoders, and language models. These methods typically replicate input visits, resulting in inadequate modeling of temporal dependencies between visits and overlooking the generation of time information, a crucial element in EHR data. Moreover, their ability to learn visit representations is limited due to simple linear mapping functions, thus compromising generation quality. To address these limitations, we propose a novel EHR data generation model called EHRPD. It is a diffusion-based model designed to predict the next visit based on the current one while also incorporating time interval estimation. To enhance generation quality and diversity, we introduce a novel time-aware visit embedding module and a pioneering predictive denoising diffusion probabilistic model (PDDPM). Additionally, we devise a predictive U-Net (PU-Net) to optimize P-DDPM.We conduct experiments on two public datasets and evaluate EHRPD from fidelity, privacy, and utility perspectives. The experimental results demonstrate the efficacy and utility of the proposed EHRPD in addressing the aforementioned limitations and advancing EHR data generation.

Automatic International Classification of Diseases (ICD) coding plays a crucial role in the extraction of relevant information from clinical notes for proper recording and billing. One of the most important directions for boosting the performance of automatic ICD coding is modeling ICD code relations. However, current methods insufficiently model the intricate relationships among ICD codes and often overlook the importance of context in clinical notes. In this paper, we propose a novel approach, a contextualized and flexible framework, to enhance the learning of ICD code representations. Our approach, unlike existing methods, employs a dependent learning paradigm that considers the context of clinical notes in modeling all possible code relations. We evaluate our approach on six public ICD coding datasets and the experimental results demonstrate the effectiveness of our approach compared to state-of-the-art baselines., Comment: LREC-Coling 2024

When legged robots physically interact with obstacles in applications such as search and rescue through rubble and planetary exploration across Martain rocks, even the most advanced ones struggle because they lack a fundamental framework to model the robot-obstacle physical interaction paralleling artificial potential fields for obstacle avoidance. To remedy this, recent studies established a novel framework - potential energy landscape modeling - that explains and predicts the destabilizing transitions across locomotor modes from the physical interaction between robots and obstacles, and governs a wide range of complex locomotion. However, this framework was confined to the laboratory because we lack methods to obtain the potential energy landscape in unknown environments. Here, we explore the feasibility of introducing this framework to such environments. We showed that a robot can reconstruct the potential energy landscape for unknown obstacles by measuring the obstacle contact forces and resulting torques. To elaborate, we developed a minimalistic robot capable of sensing contact forces and torques when propelled against a pair of grass-like obstacles. Despite the forces and torques not being fully conservative, they well-matched the potential energy landscape gradients, and the reconstructed landscape well-matched ground truth. In addition, we found that using normal forces and torques and head oscillation inspired by cockroach observations further improved the estimation of conservative ones. Our study will finally inspire free-running robots to achieve low-effort, "zero-shot" traversing clustered, large obstacles in real-world applications by sampling contact forces and torques and reconstructing the landscape around its neighboring states in real time.

Large multi-modal models (LMMs) exhibit remarkable performance across numerous tasks. However, generalist LMMs often suffer from performance degradation when tuned over a large collection of tasks. Recent research suggests that Mixture of Experts (MoE) architectures are useful for instruction tuning, but for LMMs of parameter size around O(50-100B), the prohibitive cost of replicating and storing the expert models severely limits the number of experts we can use. We propose Omni-SMoLA, an architecture that uses the Soft MoE approach to (softly) mix many multimodal low rank experts, and avoids introducing a significant number of new parameters compared to conventional MoE models. The core intuition here is that the large model provides a foundational backbone, while different lightweight experts residually learn specialized knowledge, either per-modality or multimodally. Extensive experiments demonstrate that the SMoLA approach helps improve the generalist performance across a broad range of generative vision-and-language tasks, achieving new SoTA generalist performance that often matches or outperforms single specialized LMM baselines, as well as new SoTA specialist performance.

Background: Discovering potential drug-drug interactions (DDIs) is a long-standing challenge in clinical treatments and drug developments. Recently, deep learning techniques have been developed for DDI prediction. However, they generally require a huge number of samples, while known DDIs are rare. Methods: In this work, we present KnowDDI, a graph neural network-based method that addresses the above challenge. KnowDDI enhances drug representations by adaptively leveraging rich neighborhood information from large biomedical knowledge graphs. Then, it learns a knowledge subgraph for each drug-pair to interpret the predicted DDI, where each of the edges is associated with a connection strength indicating the importance of a known DDI or resembling strength between a drug-pair whose connection is unknown. Thus, the lack of DDIs is implicitly compensated by the enriched drug representations and propagated drug similarities. Results: We evaluate KnowDDI on two benchmark DDI datasets. Results show that KnowDDI obtains the state-of-the-art prediction performance with better interpretability. We also find that KnowDDI suffers less than existing works given a sparser knowledge graph. This indicates that the propagated drug similarities play a more important role in compensating for the lack of DDIs when the drug representations are less enriched. Conclusions: KnowDDI nicely combines the efficiency of deep learning techniques and the rich prior knowledge in biomedical knowledge graphs. As an original open-source tool, KnowDDI can help detect possible interactions in a broad range of relevant interaction prediction tasks, such as protein-protein interactions, drug-target interactions and disease-gene interactions, eventually promoting the development of biomedicine and healthcare., Comment: Accepted to Nature Communications (Medicine)

Large language models (LLMs) and vision language models (VLMs) demonstrate excellent performance on a wide range of tasks by scaling up parameter counts from O(10^9) to O(10^{12}) levels and further beyond. These large scales make it impossible to adapt and deploy fully specialized models given a task of interest. Parameter-efficient fine-tuning (PEFT) emerges as a promising direction to tackle the adaptation and serving challenges for such large models. We categorize PEFT techniques into two types: intrusive and non-intrusive. Intrusive PEFT techniques directly change a model's internal architecture. Though more flexible, they introduce significant complexities for training and serving. Non-intrusive PEFT techniques leave the internal architecture unchanged and only adapt model-external parameters, such as embeddings for input. In this work, we describe AdaLink as a non-intrusive PEFT technique that achieves competitive performance compared to SoTA intrusive PEFT (LoRA) and full model fine-tuning (FT) on various tasks. We evaluate using both text-only and multimodal tasks, with experiments that account for both parameter-count scaling and training regime (with and without instruction tuning).

Personalized text generation presents a specialized mechanism for delivering content that is specific to a user's personal context. While the research progress in this area has been rapid, evaluation still presents a challenge. Traditional automated metrics such as BLEU and ROUGE primarily measure lexical similarity to human-written references, and are not able to distinguish personalization from other subtle semantic aspects, thus falling short of capturing the nuances of personalized generated content quality. On the other hand, human judgments are costly to obtain, especially in the realm of personalized evaluation. Inspired by these challenges, we explore the use of large language models (LLMs) for evaluating personalized text generation, and examine their ability to understand nuanced user context. We present AuPEL, a novel evaluation method that distills three major semantic aspects of the generated text: personalization, quality and relevance, and automatically measures these aspects. To validate the effectiveness of AuPEL, we design carefully controlled experiments and compare the accuracy of the evaluation judgments made by LLMs versus that of judgements made by human annotators, and conduct rigorous analyses of the consistency and sensitivity of the proposed metric. We find that, compared to existing evaluation metrics, AuPEL not only distinguishes and ranks models based on their personalization abilities more accurately, but also presents commendable consistency and efficiency for this task. Our work suggests that using LLMs as the evaluators of personalized text generation is superior to traditional text similarity metrics, even though interesting new challenges still remain.

Pretraining has proven to be a powerful technique in natural language processing (NLP), exhibiting remarkable success in various NLP downstream tasks. However, in the medical domain, existing pretrained models on electronic health records (EHR) fail to capture the hierarchical nature of EHR data, limiting their generalization capability across diverse downstream tasks using a single pretrained model. To tackle this challenge, this paper introduces a novel, general, and unified pretraining framework called MEDHMP, specifically designed for hierarchically multimodal EHR data. The effectiveness of the proposed MEDHMP is demonstrated through experimental results on eight downstream tasks spanning three levels. Comparisons against eighteen baselines further highlight the efficacy of our approach., Comment: Accepted by EMNLP 2023

Large Language Models (LLMs), renowned for their remarkable performance across diverse domains, present a challenge when it comes to practical deployment due to their colossal model size. In response to this challenge, efforts have been directed toward the application of traditional network pruning techniques to LLMs, uncovering a massive number of parameters that can be pruned in one-shot without hurting performance. Prevailing LLM pruning strategies have consistently adhered to the practice of uniformly pruning all layers at equivalent sparsity, resulting in robust performance. However, this observation stands in contrast to the prevailing trends observed in the field of vision models, where non-uniform layerwise sparsity typically yields stronger results. To understand the underlying reasons for this disparity, we conduct a comprehensive study and discover a strong correlation with the emergence of activation outliers in LLMs. Inspired by this finding, we introduce a novel LLM pruning methodology that incorporates a tailored set of non-uniform layerwise sparsity ratios, termed as Outlier Weighed Layerwise sparsity (OWL). The sparsity ratio of OWL is proportional to the outlier ratio observed within each layer, facilitating a more effective alignment between layerwise weight sparsity and outlier ratios. Our empirical evaluation, conducted across the LLaMA-V1 family and OPT, spanning various benchmarks, demonstrates the distinct advantages offered by OWL over previous methods. For instance, OWL exhibits a remarkable performance gain, surpassing the state-of-the-art Wanda and SparseGPT by 61.22 and 6.80 perplexity at a high sparsity level of 70%, respectively, while delivering 2.6x end-to-end inference speed-up in the DeepSparse inference engine. Codes are available at https://github.com/luuyin/OWL.

Health risk prediction is one of the fundamental tasks under predictive modeling in the medical domain, which aims to forecast the potential health risks that patients may face in the future using their historical Electronic Health Records (EHR). Researchers have developed several risk prediction models to handle the unique challenges of EHR data, such as its sequential nature, high dimensionality, and inherent noise. These models have yielded impressive results. Nonetheless, a key issue undermining their effectiveness is data insufficiency. A variety of data generation and augmentation methods have been introduced to mitigate this issue by expanding the size of the training data set through the learning of underlying data distributions. However, the performance of these methods is often limited due to their task-unrelated design. To address these shortcomings, this paper introduces a novel, end-to-end diffusion-based risk prediction model, named MedDiffusion. It enhances risk prediction performance by creating synthetic patient data during training to enlarge sample space. Furthermore, MedDiffusion discerns hidden relationships between patient visits using a step-wise attention mechanism, enabling the model to automatically retain the most vital information for generating high-quality data. Experimental evaluation on four real-world medical datasets demonstrates that MedDiffusion outperforms 14 cutting-edge baselines in terms of PR-AUC, F1, and Cohen's Kappa. We also conduct ablation studies and benchmark our model against GAN-based alternatives to further validate the rationality and adaptability of our model design. Additionally, we analyze generated data to offer fresh insights into the model's interpretability.

Molecular property prediction (MPP) plays a crucial role in biomedical applications, but it often encounters challenges due to a scarcity of labeled data. Existing works commonly adopt gradient-based strategy to update a large amount of parameters for task-level adaptation. However, the increase of adaptive parameters can lead to overfitting and poor performance. Observing that graph neural network (GNN) performs well as both encoder and predictor, we propose PACIA, a parameter-efficient GNN adapter for few-shot MPP. We design a unified adapter to generate a few adaptive parameters to modulate the message passing process of GNN. We then adopt a hierarchical adaptation mechanism to adapt the encoder at task-level and the predictor at query-level by the unified GNN adapter. Extensive results show that PACIA obtains the state-of-the-art performance in few-shot MPP problems, and our proposed hierarchical adaptation mechanism is rational and effective., Comment: 18pages

Diffusion models achieve great success in generating diverse and high-fidelity images, yet their widespread application, especially in real-time scenarios, is hampered by their inherently slow generation speed. The slow generation stems from the necessity of multi-step network inference. While some certain predictions benefit from the full computation of the model in each sampling iteration, not every iteration requires the same amount of computation, potentially leading to inefficient computation. Unlike typical adaptive computation challenges that deal with single-step generation problems, diffusion processes with a multi-step generation need to dynamically adjust their computational resource allocation based on the ongoing assessment of each step's importance to the final image output, presenting a unique set of challenges. In this work, we propose AdaDiff, an adaptive framework that dynamically allocates computation resources in each sampling step to improve the generation efficiency of diffusion models. To assess the effects of changes in computational effort on image quality, we present a timestep-aware uncertainty estimation module (UEM). Integrated at each intermediate layer, the UEM evaluates the predictive uncertainty. This uncertainty measurement serves as an indicator for determining whether to terminate the inference process. Additionally, we introduce an uncertainty-aware layer-wise loss aimed at bridging the performance gap between full models and their adaptive counterparts.

Personalized text generation is an emerging research area that has attracted much attention in recent years. Most studies in this direction focus on a particular domain by designing bespoke features or models. In this work, we propose a general approach for personalized text generation using large language models (LLMs). Inspired by the practice of writing education, we develop a multistage and multitask framework to teach LLMs for personalized generation. In writing instruction, the task of writing from sources is often decomposed into multiple steps that involve finding, evaluating, summarizing, synthesizing, and integrating information. Analogously, our approach to personalized text generation consists of multiple stages: retrieval, ranking, summarization, synthesis, and generation. In addition, we introduce a multitask setting that helps the model improve its generation ability further, which is inspired by the observation in education that a student's reading proficiency and writing ability are often correlated. We evaluate our approach on three public datasets, each of which covers a different and representative domain. Our results show significant improvements over a variety of baselines.

Automated Machine Learning (AutoML) techniques have recently been introduced to design Collaborative Filtering (CF) models in a data-specific manner. However, existing works either search architectures or hyperparameters while ignoring the fact they are intrinsically related and should be considered together. This motivates us to consider a joint hyperparameter and architecture search method to design CF models. However, this is not easy because of the large search space and high evaluation cost. To solve these challenges, we reduce the space by screening out usefulness yperparameter choices through a comprehensive understanding of individual hyperparameters. Next, we propose a two-stage search algorithm to find proper configurations from the reduced space. In the first stage, we leverage knowledge from subsampled datasets to reduce evaluation costs; in the second stage, we efficiently fine-tune top candidate models on the whole dataset. Extensive experiments on real-world datasets show better performance can be achieved compared with both hand-designed and previous searched models. Besides, ablation and case studies demonstrate the effectiveness of our search framework., Comment: Accepted by KDD 2023

Making personalized recommendation for cold-start users, who only have a few interaction histories, is a challenging problem in recommendation systems. Recent works leverage hypernetworks to directly map user interaction histories to user-specific parameters, which are then used to modulate predictor by feature-wise linear modulation function. These works obtain the state-of-the-art performance. However, the physical meaning of scaling and shifting in recommendation data is unclear. Instead of using a fixed modulation function and deciding modulation position by expertise, we propose a modulation framework called ColdNAS for user cold-start problem, where we look for proper modulation structure, including function and position, via neural architecture search. We design a search space which covers broad models and theoretically prove that this search space can be transformed to a much smaller space, enabling an efficient and robust one-shot search algorithm. Extensive experimental results on benchmark datasets show that ColdNAS consistently performs the best. We observe that different modulation functions lead to the best performance on different datasets, which validates the necessity of designing a searching-based method.

Federated learning (FL) enables multiple clients to train models collaboratively without sharing local data, which has achieved promising results in different areas, including the Internet of Things (IoT). However, end IoT devices do not have abilities to automatically annotate their collected data, which leads to the label shortage issue at the client side. To collaboratively train an FL model, we can only use a small number of labeled data stored on the server. This is a new yet practical scenario in federated learning, i.e., labels-at-server semi-supervised federated learning (SemiFL). Although several SemiFL approaches have been proposed recently, none of them can focus on the personalization issue in their model design. IoT environments make SemiFL more challenging, as we need to take device computational constraints and communication cost into consideration simultaneously. To tackle these new challenges together, we propose a novel SemiFL framework named pFedKnow. pFedKnow generates lightweight personalized client models via neural network pruning techniques to reduce communication cost. Moreover, it incorporates pretrained large models as prior knowledge to guide the aggregation of personalized client models and further enhance the framework performance. Experiment results on both image and text datasets show that the proposed pFedKnow outperforms state-of-the-art baselines as well as reducing considerable communication cost. The source code of the proposed pFedKnow is available at https://github.com/JackqqWang/pfedknow/tree/master., Comment: This paper is acceptted by SDM-2023. Jiaqi Wang and Shenglai Zeng are of equal contribution

Recent years have witnessed increasing concerns towards unfair decisions made by machine learning algorithms. To improve fairness in model decisions, various fairness notions have been proposed and many fairness-aware methods are developed. However, most of existing definitions and methods focus only on single-label classification. Fairness for multi-label classification, where each instance is associated with more than one labels, is still yet to establish. To fill this gap, we study fairness-aware multi-label classification in this paper. We start by extending Demographic Parity (DP) and Equalized Opportunity (EOp), two popular fairness notions, to multi-label classification scenarios. Through a systematic study, we show that on multi-label data, because of unevenly distributed labels, EOp usually fails to construct a reliable estimate on labels with few instances. We then propose a new framework named Similarity $s$-induced Fairness ($s_\gamma$-SimFair). This new framework utilizes data that have similar labels when estimating fairness on a particular label group for better stability, and can unify DP and EOp. Theoretical analysis and experimental results on real-world datasets together demonstrate the advantage of over existing methods $s_\gamma$-SimFair on multi-label classification tasks., Comment: AAAI2023

Time series forecasting has been a widely explored task of great importance in many applications. However, it is common that real-world time series data are recorded in a short time period, which results in a big gap between the deep model and the limited and noisy time series. In this work, we propose to address the time series forecasting problem with generative modeling and propose a bidirectional variational auto-encoder (BVAE) equipped with diffusion, denoise, and disentanglement, namely D3VAE. Specifically, a coupled diffusion probabilistic model is proposed to augment the time series data without increasing the aleatoric uncertainty and implement a more tractable inference process with BVAE. To ensure the generated series move toward the true target, we further propose to adapt and integrate the multiscale denoising score matching into the diffusion process for time series forecasting. In addition, to enhance the interpretability and stability of the prediction, we treat the latent variable in a multivariate manner and disentangle them on top of minimizing total correlation. Extensive experiments on synthetic and real-world data show that D3VAE outperforms competitive algorithms with remarkable margins. Our implementation is available at https://github.com/PaddlePaddle/PaddleSpatial/tree/main/research/D3VAE.

Large-scale Transformer models bring significant improvements for various downstream vision language tasks with a unified architecture. The performance improvements come with increasing model size, resulting in slow inference speed and increased cost for severing. While some certain predictions benefit from the full complexity of the large-scale model, not all of inputs need the same amount of computation to conduct, potentially leading to computation resource waste. To handle this challenge, early exiting is proposed to adaptively allocate computational power in term of input complexity to improve inference efficiency. The existing early exiting strategies usually adopt output confidence based on intermediate layers as a proxy of input complexity to incur the decision of skipping following layers. However, such strategies cannot apply to encoder in the widely-used unified architecture with both encoder and decoder due to difficulty of output confidence estimation in the encoder. It is suboptimal in term of saving computation power to ignore the early exiting in encoder component. To handle this challenge, we propose a novel early exiting strategy for unified visual language models, which allows dynamically skip the layers in encoder and decoder simultaneously in term of input layer-wise similarities with multiple times of early exiting, namely \textbf{MuE}. By decomposing the image and text modalities in the encoder, MuE is flexible and can skip different layers in term of modalities, advancing the inference efficiency while minimizing performance drop. Experiments on the SNLI-VE and MS COCO datasets show that the proposed approach MuE can reduce expected inference time by up to 50\% and 40\% while maintaining 99\% and 96\% performance respectively.

Standard fine-tuning of large pre-trained language models (PLMs) for downstream tasks requires updating hundreds of millions to billions of parameters, and storing a large copy of the PLM weights for every task resulting in increased cost for storing, sharing and serving the models. To address this, parameter-efficient fine-tuning (PEFT) techniques were introduced where small trainable components are injected in the PLM and updated during fine-tuning. We propose AdaMix as a general PEFT method that tunes a mixture of adaptation modules -- given the underlying PEFT method of choice -- introduced in each Transformer layer while keeping most of the PLM weights frozen. For instance, AdaMix can leverage a mixture of adapters like Houlsby or a mixture of low rank decomposition matrices like LoRA to improve downstream task performance over the corresponding PEFT methods for fully supervised and few-shot NLU and NLG tasks. Further, we design AdaMix such that it matches the same computational cost and the number of tunable parameters as the underlying PEFT method. By only tuning 0.1-0.2% of PLM parameters, we show that AdaMix outperforms SOTA parameter-efficient fine-tuning and full model fine-tuning for both NLU and NLG tasks., Comment: The paper is withdraw to avoid duplicate version of arXiv article 2205.12410. We will include new content as a updated version

While fine-tuning pre-trained networks has become a popular way to train image segmentation models, such backbone networks for image segmentation are frequently pre-trained using image classification source datasets, e.g., ImageNet. Though image classification datasets could provide the backbone networks with rich visual features and discriminative ability, they are incapable of fully pre-training the target model (i.e., backbone+segmentation modules) in an end-to-end manner. The segmentation modules are left to random initialization in the fine-tuning process due to the lack of segmentation labels in classification datasets. In our work, we propose a method that leverages Pseudo Semantic Segmentation Labels (PSSL), to enable the end-to-end pre-training for image segmentation models based on classification datasets. PSSL was inspired by the observation that the explanation results of classification models, obtained through explanation algorithms such as CAM, SmoothGrad and LIME, would be close to the pixel clusters of visual objects. Specifically, PSSL is obtained for each image by interpreting the classification results and aggregating an ensemble of explanations queried from multiple classifiers to lower the bias caused by single models. With PSSL for every image of ImageNet, the proposed method leverages a weighted segmentation learning procedure to pre-train the segmentation network en masse. Experiment results show that, with ImageNet accompanied by PSSL as the source dataset, the proposed end-to-end pre-training strategy successfully boosts the performance of various segmentation models, i.e., PSPNet-ResNet50, DeepLabV3-ResNet50, and OCRNet-HRNetW18, on a number of segmentation tasks, such as CamVid, VOC-A, VOC-C, ADE20K, and CityScapes, with significant improvements. The source code is availabel at https://github.com/PaddlePaddle/PaddleSeg., Comment: Accepted by Machine Learning

Nonconvex regularization has been popularly used in low-rank matrix learning. However, extending it for low-rank tensor learning is still computationally expensive. To address this problem, we develop an efficient solver for use with a nonconvex extension of the overlapped nuclear norm regularizer. Based on the proximal average algorithm, the proposed algorithm can avoid expensive tensor folding/unfolding operations. A special "sparse plus low-rank" structure is maintained throughout the iterations, and allows fast computation of the individual proximal steps. Empirical convergence is further improved with the use of adaptive momentum. We provide convergence guarantees to critical points on smooth losses and also on objectives satisfying the Kurdyka-{\L}ojasiewicz condition. While the optimization problem is nonconvex and nonsmooth, we show that its critical points still have good statistical performance on the tensor completion problem. Experiments on various synthetic and real-world data sets show that the proposed algorithm is efficient in both time and space and more accurate than the existing state-of-the-art., Comment: Accepted to JMLR in 2022

To traverse complex natural terrain, animals often transition between locomotor modes. It is well known that locomotor transitions can be induced by switching in neural control circuits or be driven by a need to minimize metabolic energetic cost. Recent work discovered that locomotor transitions in complex 3-D terrain cluttered with large obstacles can also emerge from physical interaction with the environment controlled by the nervous system. To traverse cluttered, stiff grass-like beams, the discoid cockroach often transitions from using a strenuous pitch mode to push across to using a less strenuous roll mode to maneuver through the gaps, during which a potential energy barrier must be overcome. Although previous robotic physical modeling demonstrated that kinetic energy fluctuation from body oscillation generated by leg propulsion can help overcome the barrier and facilitate this transition, the animal was observed to transition even when the barrier still exceeds kinetic energy fluctuation. Here, we further studied whether and how the cockroach makes active adjustments to facilitate this locomotor transition to traverse cluttered beams. We observed that the animal flexed its head and abdomen, reduced hind leg sprawl, and used both hind legs differentially during the pitch-to-roll transition, which were absent when running on a flat ground. Using a refined potential energy landscape with additional degrees of freedom modeling these adjustments, we found that head flexion did not substantially reduce the transition barrier, whereas the leg sprawl reduction did so dramatically. We discussed likely functions of the observed adjustments and suggested future directions.

Robots excel at avoiding obstacles but struggle to traverse complex 3-D terrain with cluttered large obstacles. By contrast, insects like cockroaches excel at doing so. Recent research in our lab elucidated how locomotor transitions emerge from locomotor-environment interaction for diverse locomotor challenges abstracted from complex 3-D terrain and the strategies to overcome them. Here we built on these fundamental insights to develop a cockroach-inspired legged robot, Omni-Roach, that integrated these strategies to achieve multi-modal locomotion and provide a robophysical model to study the trade-off between multi-functionality and performance. The robot was based on the RHex design with six compliant legs and featured a rounded body with two wings that can open and a tail with pitch and yaw degrees of freedom. After two development and testing iterations, our robot was capable of overcoming all locomotor challenges with a high performance and success rate. It traversed cluttered rigid pillars only 1.1x robot body width apart, a 2.5x hip height bump, a 0.75x body length gap, densely cluttered flexible beams only 65% body width apart, and self-righted within 4 seconds. Systematic beam traversal experiments further revealed that a downward-pointing tail oscillating laterally helps roll the body into beam gaps and break frictional and interlocking contact to traverse. Our work highlights the usefulness of multi-functional appendages and exaptation for large obstacle traversal.

Short text classification is a fundamental task in natural language processing. It is hard due to the lack of context information and labeled data in practice. In this paper, we propose a new method called SHINE, which is based on graph neural network (GNN), for short text classification. First, we model the short text dataset as a hierarchical heterogeneous graph consisting of word-level component graphs which introduce more semantic and syntactic information. Then, we dynamically learn a short document graph that facilitates effective label propagation among similar short texts. Thus, compared with existing GNN-based methods, SHINE can better exploit interactions between nodes of the same types and capture similarities between short texts. Extensive experiments on various benchmark short text datasets show that SHINE consistently outperforms state-of-the-art methods, especially with fewer labels., Comment: Accepted to EMNLP 2021