Fine-tuning pretrained language models (PLMs) for downstream tasks is a large-scale optimization problem, in which the choice of the training algorithm critically determines how well the trained model can generalize to unseen test data, especially in the context of few-shot learning. To achieve good generalization performance and avoid overfitting, techniques such as data augmentation and pruning are often applied. However, adding these regularizations necessitates heavy tuning of the hyperparameters of optimization algorithms, such as the popular Adam optimizer. In this paper, we propose a two-stage fine-tuning method, PAC-tuning, to address this optimization challenge. First, based on PAC-Bayes training, PAC-tuning directly minimizes the PAC-Bayes generalization bound to learn proper parameter distribution. Second, PAC-tuning modifies the gradient by injecting noise with the variance learned in the first stage into the model parameters during training, resulting in a variant of perturbed gradient descent (PGD). In the past, the few-shot scenario posed difficulties for PAC-Bayes training because the PAC-Bayes bound, when applied to large models with limited training data, might not be stringent. Our experimental results across 5 GLUE benchmark tasks demonstrate that PAC-tuning successfully handles the challenges of fine-tuning tasks and outperforms strong baseline methods by a visible margin, further confirming the potential to apply PAC training for any other settings where the Adam optimizer is currently used for training.
The existing research on robust Graph Neural Networks (GNNs) fails to acknowledge the significance of directed graphs in providing rich information about networks' inherent structure. This work presents the first investigation into the robustness of GNNs in the context of directed graphs, aiming to harness the profound trust implications offered by directed graphs to bolster the robustness and resilience of GNNs. Our study reveals that existing directed GNNs are not adversarially robust. In pursuit of our goal, we introduce a new and realistic directed graph attack setting and propose an innovative, universal, and efficient message-passing framework as a plug-in layer to significantly enhance the robustness of GNNs. Combined with existing defense strategies, this framework achieves outstanding clean accuracy and state-of-the-art robust performance, offering superior defense against both transfer and adaptive attacks. The findings in this study reveal a novel and promising direction for this crucial research area. The code will be made publicly available upon the acceptance of this work.
It is widely recognized that the generalization ability of neural networks can be greatly enhanced through carefully designing the training procedure. The current state-of-the-art training approach involves utilizing stochastic gradient descent (SGD) or Adam optimization algorithms along with a combination of additional regularization techniques such as weight decay, dropout, or noise injection. Optimal generalization can only be achieved by tuning a multitude of hyperparameters through grid search, which can be time-consuming and necessitates additional validation datasets. To address this issue, we introduce a practical PAC-Bayes training framework that is nearly tuning-free and requires no additional regularization while achieving comparable testing performance to that of SGD/Adam after a complete grid search and with extra regularizations. Our proposed algorithm demonstrates the remarkable potential of PAC training to achieve state-of-the-art performance on deep neural networks with enhanced robustness and interpretability.
It is well known that the finite step-size ($h$) in Gradient Descent (GD) implicitly regularizes solutions to flatter minima. A natural question to ask is "Does the momentum parameter $\beta$ play a role in implicit regularization in Heavy-ball (H.B) momentum accelerated gradient descent (GD+M)?". To answer this question, first, we show that the discrete H.B momentum update (GD+M) follows a continuous trajectory induced by a modified loss, which consists of an original loss and an implicit regularizer. Then, we show that this implicit regularizer for (GD+M) is stronger than that of (GD) by factor of $(\frac{1+\beta}{1-\beta})$, thus explaining why (GD+M) shows better generalization performance and higher test accuracy than (GD). Furthermore, we extend our analysis to the stochastic version of gradient descent with momentum (SGD+M) and characterize the continuous trajectory of the update of (SGD+M) in a pointwise sense. We explore the implicit regularization in (SGD+M) and (GD+M) through a series of experiments validating our theory.
Monitoring changes inside a reservoir in real time is crucial for the success of CO2 injection and long-term storage. Machine learning (ML) is well-suited for real-time CO2 monitoring because of its computational efficiency. However, most existing applications of ML yield only one prediction (i.e., the expectation) for a given input, which may not properly reflect the distribution of the testing data, if it has a shift with respect to that of the training data. The Simultaneous Quantile Regression (SQR) method can estimate the entire conditional distribution of the target variable of a neural network via pinball loss. Here, we incorporate this technique into seismic inversion for purposes of CO2 monitoring. The uncertainty map is then calculated pixel by pixel from a particular prediction interval around the median. We also propose a novel data-augmentation method by sampling the uncertainty to further improve prediction accuracy. The developed methodology is tested on synthetic Kimberlina data, which are created by the Department of Energy and based on a CO2 capture and sequestration (CCS) project in California. The results prove that the proposed network can estimate the subsurface velocity rapidly and with sufficient resolution. Furthermore, the computed uncertainty quantifies the prediction accuracy. The method remains robust even if the testing data are distorted due to problems in the field data acquisition. Another test demonstrates the effectiveness of the developed data-augmentation method in increasing the spatial resolution of the estimated velocity field and in reducing the prediction error.
Signed networks are ubiquitous in many real-world applications (e.g., social networks encoding trust/distrust relationships, correlation networks arising from time series data). While many signed networks are directed, there is a lack of survey papers and software packages on graph neural networks (GNNs) specially designed for directed networks. In this paper, we present PyTorch Geometric Signed Directed, a survey and software on GNNs for signed and directed networks. We review typical tasks, loss functions and evaluation metrics in the analysis of signed and directed networks, discuss data used in related experiments, and provide an overview of methods proposed. The deep learning framework consists of easy-to-use GNN models, synthetic and real-world data, as well as task-specific evaluation metrics and loss functions for signed and directed networks. The software is presented in a modular fashion, so that signed and directed networks can also be treated separately. As an extension library for PyTorch Geometric, our proposed software is maintained with open-source releases, detailed documentation, continuous integration, unit tests and code coverage checks. Our code is publicly available at \url{https://github.com/SherylHYX/pytorch_geometric_signed_directed}.
Multi-physical inversion plays a critical role in geophysics. It has been widely used to infer various physical properties (such as velocity and conductivity), simultaneously. Among those inversion problems, some are explicitly governed by partial differential equations (PDEs), while others are not. Without explicit governing equations, conventional multi-physical inversion techniques will not be feasible and data-driven inversion require expensive full labels. To overcome this issue, we develop a new data-driven multi-physics inversion technique with extremely weak supervision. Our key finding is that the pseudo labels can be constructed by learning the local relationship among geophysical properties at very sparse locations. We explore a multi-physics inversion problem from two distinct measurements (seismic and EM data) to three geophysical properties (velocity, conductivity, and CO$_2$ saturation). Our results show that we are able to invert for properties without explicit governing equations. Moreover, the label data on three geophysical properties can be significantly reduced by 50 times (from 100 down to only 2 locations).
We present OpenFWI, a collection of large-scale open-source benchmark datasets for seismic full waveform inversion (FWI). OpenFWI is the first-of-its-kind in the geoscience and machine learning community to facilitate diversified, rigorous, and reproducible research on machine learning-based FWI. OpenFWI includes datasets of multiple scales, encompasses diverse domains, and covers various levels of model complexity. Along with the dataset, we also perform an empirical study on each dataset with a fully-convolutional deep learning model. OpenFWI has been meticulously maintained and will be regularly updated with new data and experimental results. We appreciate the inputs from the community to help us further improve OpenFWI. At the current version, we publish seven datasets in OpenFWI, of which one is specified for 3D FWI and the rest are for 2D scenarios. All datasets and related information can be accessed through our website at https://openfwi.github.io/.
This paper investigates unsupervised learning of Full-Waveform Inversion (FWI), which has been widely used in geophysics to estimate subsurface velocity maps from seismic data. This problem is mathematically formulated by a second order partial differential equation (PDE), but is hard to solve. Moreover, acquiring velocity map is extremely expensive, making it impractical to scale up a supervised approach to train the mapping from seismic data to velocity maps with convolutional neural networks (CNN). We address these difficulties by integrating PDE and CNN in a loop, thus shifting the paradigm to unsupervised learning that only requires seismic data. In particular, we use finite difference to approximate the forward modeling of PDE as a differentiable operator (from velocity map to seismic data) and model its inversion by CNN (from seismic data to velocity map). Hence, we transform the supervised inversion task into an unsupervised seismic data reconstruction task. We also introduce a new large-scale dataset OpenFWI, to establish a more challenging benchmark for the community. Experiment results show that our model (using seismic data alone) yields comparable accuracy to the supervised counterpart (using both seismic data and velocity map). Furthermore, it outperforms the supervised model when involving more seismic data.
Deep learning and data-driven approaches have shown great potential in scientific domains. The promise of data-driven techniques relies on the availability of a large volume of high-quality training datasets. Due to the high cost of obtaining data through expensive physical experiments, instruments, and simulations, data augmentation techniques for scientific applications have emerged as a new direction for obtaining scientific data recently. However, existing data augmentation techniques originating from computer vision, yield physically unacceptable data samples that are not helpful for the domain problems that we are interested in. In this paper, we develop new physics-informed data augmentation techniques based on convolutional neural networks. Specifically, our generative models leverage different physics knowledge (such as governing equations, observable perception, and physics phenomena) to improve the quality of the synthetic data. To validate the effectiveness of our data augmentation techniques, we apply them to solve a subsurface seismic full-waveform inversion using simulated CO$_2$ leakage data. Our interest is to invert for subsurface velocity models associated with very small CO$_2$ leakage. We validate the performance of our methods using comprehensive numerical tests. Via comparison and analysis, we show that data-driven seismic imaging can be significantly enhanced by using our physics-informed data augmentation techniques. Particularly, the imaging quality has been improved by 15% in test scenarios of general-sized leakage and 17% in small-sized leakage when using an augmented training set obtained with our techniques.