Graph representation learning has attracted a surge of interest recently, whose target at learning discriminant embedding for each node in the graph. Most of these representation methods focus on supervised learning and heavily depend on label information. However, annotating graphs are expensive to obtain in the real world, especially in specialized domains (i.e. biology), as it needs the annotator to have the domain knowledge to label the graph. To approach this problem, self-supervised learning provides a feasible solution for graph representation learning. In this paper, we propose a Multi-Level Graph Contrastive Learning (MLGCL) framework for learning robust representation of graph data by contrasting space views of graphs. Specifically, we introduce a novel contrastive view - topological and feature space views. The original graph is first-order approximation structure and contains uncertainty or error, while the $k$NN graph generated by encoding features preserves high-order proximity. Thus $k$NN graph generated by encoding features not only provide a complementary view, but is more suitable to GNN encoder to extract discriminant representation. Furthermore, we develop a multi-level contrastive mode to preserve the local similarity and semantic similarity of graph-structured data simultaneously. Extensive experiments indicate MLGCL achieves promising results compared with the existing state-of-the-art graph representation learning methods on seven datasets.

This paper proposes an invariant causal predictor that is robust to distribution shift across domains and maximally reserves the transferable invariant information. Based on a disentangled causal factorization, we formulate the distribution shift as soft interventions in the system, which covers a wide range of cases for distribution shift as we do not make prior specifications on the causal structure or the intervened variables. Instead of imposing regularizations to constrain the invariance of the predictor, we propose to predict by the intervened conditional expectation based on the do-operator and then prove that it is invariant across domains. More importantly, we prove that the proposed predictor is the robust predictor that minimizes the worst-case quadratic loss among the distributions of all domains. For empirical learning, we propose an intuitive and flexible estimating method based on data regeneration and present a local causal discovery procedure to guide the regeneration step. The key idea is to regenerate data such that the regenerated distribution is compatible with the intervened graph, which allows us to incorporate standard supervised learning methods with the regenerated data. Experimental results on both synthetic and real data demonstrate the efficacy of our predictor in improving the predictive accuracy and robustness across domains.

Interconnected road lanes are a central concept for navigating urban roads. Currently, most autonomous vehicles rely on preconstructed lane maps as designing an algorithmic model is difficult. However, the generation and maintenance of such maps is costly and hinders large-scale adoption of autonomous vehicle technology. This paper presents the first self-supervised learning method to train a model to infer a spatially grounded lane-level road network graph based on a dense segmented representation of the road scene generated from onboard sensors. A formal road lane network model is presented and proves that any structured road scene can be represented by a directed acyclic graph of at most depth three while retaining the notion of intersection regions, and that this is the most compressed representation. The formal model is implemented by a hybrid neural and search-based model, utilizing a novel barrier function loss formulation for robust learning from partial labels. Experiments are conducted for all common road intersection layouts. Results show that the model can generalize to new road layouts, unlike previous approaches, demonstrating its potential for real-world application as a practical learning-based lane-level map generator.

For artificial learning systems, continual learning over time from a stream of data is essential. The burgeoning studies on supervised continual learning have achieved great progress, while the study of catastrophic forgetting in unsupervised learning is still blank. Among unsupervised learning methods, self-supervise learning method shows tremendous potential on visual representation without any labeled data at scale. To improve the visual representation of self-supervised learning, larger and more varied data is needed. In the real world, unlabeled data is generated at all times. This circumstance provides a huge advantage for the learning of the self-supervised method. However, in the current paradigm, packing previous data and current data together and training it again is a waste of time and resources. Thus, a continual self-supervised learning method is badly needed. In this paper, we make the first attempt to implement the continual contrastive self-supervised learning by proposing a rehearsal method, which keeps a few exemplars from the previous data. Instead of directly combining saved exemplars with the current data set for training, we leverage self-supervised knowledge distillation to transfer contrastive information among previous data to the current network by mimicking similarity score distribution inferred by the old network over a set of saved exemplars. Moreover, we build an extra sample queue to assist the network to distinguish between previous and current data and prevent mutual interference while learning their own feature representation. Experimental results show that our method performs well on CIFAR100 and ImageNet-Sub. Compared with the baselines, which learning tasks without taking any technique, we improve the image classification top-1 accuracy by 1.60% on CIFAR100, 2.86% on ImageNet-Sub and 1.29% on ImageNet-Full under 10 incremental steps setting.

Self-Supervised Learning (SSL) is a pre-training alternative to transfer learning. Even though SSL emerged from massive NLP datasets, it has also shown significant progress in computer vision. Self-supervised learning in computer vision started from pretext tasks like rotation, jigsaw puzzles or even video ordering. All of these methods were formulating hand-crafted classification problems to generate labels without human annotators. Because many application domains are deprived of human labels. To this end, self-supervised learning is one way to transfer weights. By pretraining your model on labels that are artificially produced from the data.

Like EC2 instances, Amazon offers a variety of instances for training in SageMaker. Based on CPU cores, memory size and presence of GPUs, they come with different on-demand prices. A complete list can be found at Amazon SageMaker Pricing Page. From my point of view, too many choices is as bad as having none or limited choices, when you don't have a clear idea how to choose. How do you handle the overwhelm of choosing an appropriate instance type for your training task on SageMaker?

Despite pre-trained language models have proven useful for learning high-quality semantic representations, these models are still vulnerable to simple perturbations. Recent works aimed to improve the robustness of pre-trained models mainly focus on adversarial training from perturbed examples with similar semantics, neglecting the utilization of different or even opposite semantics. Different from the image processing field, the text is discrete and few word substitutions can cause significant semantic changes. To study the impact of semantics caused by small perturbations, we conduct a series of pilot experiments and surprisingly find that adversarial training is useless or even harmful for the model to detect these semantic changes. To address this problem, we propose Contrastive Learning with semantIc Negative Examples (CLINE), which constructs semantic negative examples unsupervised to improve the robustness under semantically adversarial attacking. By comparing with similar and opposite semantic examples, the model can effectively perceive the semantic changes caused by small perturbations. Empirical results show that our approach yields substantial improvements on a range of sentiment analysis, reasoning, and reading comprehension tasks. And CLINE also ensures the compactness within the same semantics and separability across different semantics in sentence-level.

Few-shot learning (FSL) aims to train a strong classifier using limited labeled examples. Many existing works take the meta-learning approach, sampling few-shot tasks in turn and optimizing the few-shot learner's performance on classifying the query examples. In this paper, we point out two potential weaknesses of this approach. First, the sampled query examples may not provide sufficient supervision for the few-shot learner. Second, the effectiveness of meta-learning diminishes sharply with increasing shots (i.e., the number of training examples per class). To resolve these issues, we propose a novel objective to directly train the few-shot learner to perform like a strong classifier. Concretely, we associate each sampled few-shot task with a strong classifier, which is learned with ample labeled examples. The strong classifier has a better generalization ability and we use it to supervise the few-shot learner. We present an efficient way to construct the strong classifier, making our proposed objective an easily plug-and-play term to existing meta-learning based FSL methods. We validate our approach in combinations with many representative meta-learning methods. On several benchmark datasets including miniImageNet and tiredImageNet, our approach leads to a notable improvement across a variety of tasks. More importantly, with our approach, meta-learning based FSL methods can consistently outperform non-meta-learning based ones, even in a many-shot setting, greatly strengthening their applicability.

High levels of sparsity and strong class imbalance are ubiquitous challenges that are often presented simultaneously in real-world time series data. While most methods tackle each problem separately, our proposed approach handles both in conjunction, while imposing fewer assumptions on the data. In this work, we propose leveraging a self-supervised learning method, specifically Autoregressive Predictive Coding (APC), to learn relevant hidden representations of time series data in the context of both missing data and class imbalance. We apply APC using either a GRU or GRU-D encoder on two real-world datasets, and show that applying one-step-ahead prediction with APC improves the classification results in all settings. In fact, by applying GRU-D - APC, we achieve state-of-the-art AUPRC results on the Physionet benchmark.

Recently, pre-training has been a hot topic in Computer Vision (and also NLP), especially one of the breakthroughs in NLP -- BERT, which proposed a method to train an NLP model by using a "self-supervised" signal. In short, we come up with an algorithm that can generate a "pseudo-label" itself (meaning a label that is true for a specific task), then we treat the learning task as a supervised learning task with the generated pseudo-label. It is commonly called "Pretext Task". For example, BERT uses mask word prediction to train the model (we can then say it is a pre-trained model after it is trained), then fine-tune the model with the task we want (usually called "Downstream Task"), e.g. The mask word prediction is to randomly mask a word in the sentence, and ask the model to predict what is that word given the sentence.