Existing self-supervised techniques have extreme computational requirements and suffer a substantial drop in performance with a reduction in batch size or pretraining epochs. This paper presents Cross Architectural - Self Supervision (CASS), a novel self-supervised learning approach that leverages Transformer and CNN simultaneously. Compared to the existing state-of-the-art self-supervised learning approaches, we empirically show that CASS-trained CNNs and Transformers across four diverse datasets gained an average of 3.8% with 1% labeled data, 5.9% with 10% labeled data, and 10.13% with 100% labeled data while taking 69% less time. We also show that CASS is much more robust to changes in batch size and training epochs than existing state-of-the-art self-supervised learning approaches. We have open-sourced our code at https://github.com/pranavsinghps1/CASS.

This paper focuses on the data augmentation for low-resource NLP tasks where the training set is limited. The existing solutions either leverage task-independent heuristic rules (e.g., Synonym Replacement) or fine-tune general-purpose pre-trained language models (e.g., GPT2) using the limited training instances to produce new synthetic data. Consequently, they have trivial task-specific knowledge and are limited to yielding low-quality synthetic data. To combat this issue, we propose Knowledge Mixture Data Augmentation Model (KnowDA) which is an Seq2Seq language model pre-trained on a mixture of diverse NLP tasks under a novel framework of Knowledge Mixture Training (KoMT). The goal of KoMT is to condense diverse NLP task-specific knowledge into the single KnowDA model (i.e., all-in-one) such that KnowDA could utilize these knowledge to quickly grasp the inherent synthesis law of the target task through limited training instances. Specifically, KoMT reformulates input examples from various heterogeneous NLP tasks into a unified text-to-text format, and employs denoising training objectives in different granularity to learn to reconstruct partial or complete samples. To the best of our knowledge, we are the first attempt to apply 100+ NLP multi-task training for data augmentation. Extensive experiments show that i) the synthetic data produced by KnowDA successfully improves performance of the strong pre-trained language models (i.e., Bert, ALBert and Deberta) by a large margin on the low-resource NLP benchmark FewGLUE, CoNLL'03 and WikiAnn; ii) KnowDA successfully transfers the task knowledge to NLP tasks whose types are seen and unseen in KoMT.

Despite rapid advancements in lifelong learning (LLL) research, a large body of research mainly focuses on improving the performance in the existing \textit{static} continual learning (CL) setups. These methods lack the ability to succeed in a rapidly changing \textit{dynamic} environment, where an AI agent needs to quickly learn new instances in a `single pass' from the non-i.i.d (also possibly temporally contiguous/coherent) data streams without suffering from catastrophic forgetting. For practical applicability, we propose a novel lifelong learning approach, which is streaming, i.e., a single input sample arrives in each time step, single pass, class-incremental, and subject to be evaluated at any moment. To address this challenging setup and various evaluation protocols, we propose a Bayesian framework, that enables fast parameter update, given a single training example, and enables any-time inference. We additionally propose an implicit regularizer in the form of snap-shot self-distillation, which effectively minimizes the forgetting further. We further propose an effective method that efficiently selects a subset of samples for online memory rehearsal and employs a new replay buffer management scheme that significantly boosts the overall performance. Our empirical evaluations and ablations demonstrate that the proposed method outperforms the prior works by large margins.

Using techniques of Symbolic Computation and Computer Algebra, we apply the concept of persistent homology to obtain a new semi-supervised learning method. Machine Learning and Deep Learning methods have become the state-of-the-art approach for solving data classification tasks. In order to use those methods, it is necessary to acquire and label a considerable amount of data; however, this is not straightforward in some fields, since data annotation is time consuming and may require expert knowledge. This challenge can be tackled by means of semi-supervised learning methods that take advantage of both labelled and unlabelled data. In our team we have applied this Machine Learning paradigm in various applied projects (e.g.

Conventional rule learning algorithms aim at finding a set of simple rules, where each rule covers as many examples as possible. In this paper, we argue that the rules found in this way may not be the optimal explanations for each of the examples they cover. Instead, we propose an efficient algorithm that aims at finding the best rule covering each training example in a greedy optimization consisting of one specialization and one generalization loop. These locally optimal rules are collected and then filtered for a final rule set, which is much larger than the sets learned by conventional rule learning algorithms. A new example is classified by selecting the best among the rules that cover this example. In our experiments on small to very large datasets, the approach's average classification accuracy is higher than that of state-of-the-art rule learning algorithms. Moreover, the algorithm is highly efficient and can inherently be processed in parallel without affecting the learned rule set and so the classification accuracy. We thus believe that it closes an important gap for large-scale classification rule induction.

Influence function, a method from robust statistics, measures the changes of model parameters or some functions about model parameters concerning the removal or modification of training instances. It is an efficient and useful post-hoc method for studying the interpretability of machine learning models without the need for expensive model re-training. Recently, graph convolution networks (GCNs), which operate on graph data, have attracted a great deal of attention. However, there is no preceding research on the influence functions of GCNs to shed light on the effects of removing training nodes/edges from an input graph. Since the nodes/edges in a graph are interdependent in GCNs, it is challenging to derive influence functions for GCNs. To fill this gap, we started with the simple graph convolution (SGC) model that operates on an attributed graph and formulated an influence function to approximate the changes in model parameters when a node or an edge is removed from an attributed graph. Moreover, we theoretically analyzed the error bound of the estimated influence of removing an edge. We experimentally validated the accuracy and effectiveness of our influence estimation function. In addition, we showed that the influence function of an SGC model could be used to estimate the impact of removing training nodes/edges on the test performance of the SGC without re-training the model. Finally, we demonstrated how to use influence functions to guide the adversarial attacks on GCNs effectively.

Mental disorders such as depression and suicidal ideation are hazardous, affecting more than 300 million people over the world. However, on social media, mental disorder symptoms can be observed, and automated approaches are increasingly capable of detecting them. The considerable number of social media users and the tremendous quantity of user-generated data on social platforms provide a unique opportunity for researchers to distinguish patterns that correlate with mental status. This research offers a roadmap for analysis, where mental state detection can be based on machine learning techniques. We describe the common approaches for predicting and identifying the disorder using user-generated content. This research is organized according to the data collection, feature extraction, and prediction algorithms. Furthermore, we review several recent studies conducted to explore different features of candidate profiles and their analytical methods. Following, we debate various aspects of the development of experimental auto-detection frameworks for identifying users who suffer from disorders, and we conclude with a discussion of future trends. The introduced methods can help complement screening procedures, identify at-risk people through social media monitoring on a large scale, and make disorders easier to treat in the future.

In this paper, we prove that separable negative log-likelihood losses for structured prediction are not necessarily Bayes consistent, or, in other words, minimizing these losses may not result in a model that predicts the most probable structure in the data distribution for a given input. This fact opens the question of whether these losses are well-adapted for structured prediction and, if so, why.

Backward compatibility of model predictions is a desired property when updating a machine learning driven application. It allows to seamlessly improve the underlying model without introducing regression bugs. In classification tasks these bugs occur in the form of negative flips. This means an instance that was correctly classified by the old model is now classified incorrectly by the updated model. This has direct negative impact on the user experience of such systems e.g. a frequently used voice assistant query is suddenly misclassified. A common reason to update the model is when new training data becomes available and needs to be incorporated. Simply retraining the model with the updated data introduces the unwanted negative flips. We study the problem of regression during data updates and propose Backward Compatible Weight Interpolation (BCWI). This method interpolates between the weights of the old and new model and we show in extensive experiments that it reduces negative flips without sacrificing the improved accuracy of the new model. BCWI is straight forward to implement and does not increase inference cost. We also explore the use of importance weighting during interpolation and averaging the weights of multiple new models in order to further reduce negative flips.

Self-supervised learning (SSL) methods aim to exploit the abundance of unlabelled data for machine learning (ML), however the underlying principles are often method-specific. An SSL framework derived from biological first principles of embodied learning could unify the various SSL methods, help elucidate learning in the brain, and possibly improve ML. SSL commonly transforms each training datapoint into a pair of views, uses the knowledge of this pairing as a positive (i.e. non-contrastive) self-supervisory sign, and potentially opposes it to unrelated, (i.e. contrastive) negative examples. Here, we show that this type of self-supervision is an incomplete implementation of a concept from neuroscience, the Efference Copy (EC). Specifically, the brain also transforms the environment through efference, i.e. motor commands, however it sends to itself an EC of the full commands, i.e. more than a mere SSL sign. In addition, its action representations are likely egocentric. From such a principled foundation we formally recover and extend SSL methods such as SimCLR, BYOL, and ReLIC under a common theoretical framework, i.e. Self-supervision Through Efference Copies (S-TEC). Empirically, S-TEC restructures meaningfully the within- and between-class representations. This manifests as improvement in recent strong SSL baselines in image classification, segmentation, object detection, and in audio. These results hypothesize a testable positive influence from the brain's motor outputs onto its sensory representations.