Unsupervised learning is argued to be the dark matter of human intelligence. To build in this direction, this paper focuses on unsupervised learning from an abundance of unlabeled data followed by few-shot fine-tuning on a downstream classification task. To this aim, we extend a recent study on adopting contrastive learning for self-supervised pre-training by incorporating class-level cognizance through iterative clustering and re-ranking and by expanding the contrastive optimization loss to account for it. To our knowledge, our experimentation both in standard and cross-domain scenarios demonstrate that we set a new state-of-the-art (SoTA) in (5-way, 1 and 5-shot) settings of standard mini-ImageNet benchmark as well as the (5-way, 5 and 20-shot) settings of cross-domain CDFSL benchmark. Our code and experimentation can be found in our GitHub repository: https://github.com/ojss/c3lr.

The accelerated advancements in the data science field in the last few decades has benefitted many other fields including Structural Health Monitoring (SHM). Particularly, the employment of Artificial Intelligence (AI) such as Machine Learning (ML) and Deep Learning (DL) methods towards vibration-based damage diagnostics of civil structures have seen a great interest due to their nature of supreme performance in learning from data. Along with diagnostics, damage prognostics also hold a vital prominence, such as estimating the remaining useful life of civil structures. Currently used AI-based data-driven methods for damage diagnostics and prognostics are centered on historical data of the structures and require a substantial amount of data to directly form the prediction models. Although some of these methods are generative-based models, after learning the distribution of the data, they are used to perform ML or DL tasks such as classification, regression, clustering, etc. In this study, a variant of Generative Adversarial Networks (GAN), Cycle-Consistent Wasserstein Deep Convolutional GAN with Gradient Penalty (CycleWDCGAN-GP) model is used to answer some of the most important questions in SHM: "How does the dynamic signature of a structure transition from undamaged to damaged conditions?" and "What is the nature of such transition?". The outcomes of this study demonstrate that the proposed model can accurately generate the possible future responses of a structure for potential future damaged conditions. In other words, with the proposed methodology, the stakeholders will be able to understand the damaged condition of structures while the structures are still in healthy (undamaged) conditions. This tool will enable them to be more proactive in overseeing the life cycle performance of structures as well as assist in remaining useful life predictions.

Mobile Crowdsensing systems are vulnerable to various attacks as they build on non-dedicated and ubiquitous properties. Machine learning (ML)-based approaches are widely investigated to build attack detection systems and ensure MCS systems security. However, adversaries that aim to clog the sensing front-end and MCS back-end leverage intelligent techniques, which are challenging for MCS platform and service providers to develop appropriate detection frameworks against these attacks. Generative Adversarial Networks (GANs) have been applied to generate synthetic samples, that are extremely similar to the real ones, deceiving classifiers such that the synthetic samples are indistinguishable from the originals. Previous works suggest that GAN-based attacks exhibit more crucial devastation than empirically designed attack samples, and result in low detection rate at the MCS platform. With this in mind, this paper aims to detect intelligently designed illegitimate sensing service requests by integrating a GAN-based model. To this end, we propose a two-level cascading classifier that combines the GAN discriminator with a binary classifier to prevent adversarial fake tasks. Through simulations, we compare our results to a single-level binary classifier, and the numeric results show that proposed approach raises Adversarial Attack Detection Rate (AADR), from $0\%$ to $97.5\%$ by KNN/NB, from $45.9\%$ to $100\%$ by Decision Tree. Meanwhile, with two-levels classifiers, Original Attack Detection Rate (OADR) improves for the three binary classifiers, with comparison, such as NB from $26.1\%$ to $61.5\%$.

Learning from unlabeled or partially labeled data to alleviate human labeling remains a challenging research topic in 3D modeling. Along this line, unsupervised representation learning is a promising direction to auto-extract features without human intervention. This paper proposes a general unsupervised approach, named \textbf{ConClu}, to perform the learning of point-wise and global features by jointly leveraging point-level clustering and instance-level contrasting. Specifically, for one thing, we design an Expectation-Maximization (EM) like soft clustering algorithm that provides local supervision to extract discriminating local features based on optimal transport. We show that this criterion extends standard cross-entropy minimization to an optimal transport problem, which we solve efficiently using a fast variant of the Sinkhorn-Knopp algorithm. For another, we provide an instance-level contrasting method to learn the global geometry, which is formulated by maximizing the similarity between two augmentations of one point cloud. Experimental evaluations on downstream applications such as 3D object classification and semantic segmentation demonstrate the effectiveness of our framework and show that it can outperform state-of-the-art techniques.

The recent proposed self-supervised learning (SSL) approaches successfully demonstrate the great potential of supplementing learning algorithms with additional unlabeled data. However, it is still unclear whether the existing SSL algorithms can fully utilize the information of both labelled and unlabeled data. This paper gives an affirmative answer for the reconstruction-based SSL algorithm \citep{lee2020predicting} under several statistical models. While existing literature only focuses on establishing the upper bound of the convergence rate, we provide a rigorous minimax analysis, and successfully justify the rate-optimality of the reconstruction-based SSL algorithm under different data generation models. Furthermore, we incorporate the reconstruction-based SSL into the existing adversarial training algorithms and show that learning from unlabeled data helps improve the robustness.

Model identifiability is a desirable property in the context of unsupervised representation learning. In absence thereof, different models may be observationally indistinguishable while yielding representations that are nontrivially related to one another, thus making the recovery of a ground truth generative model fundamentally impossible, as often shown through suitably constructed counterexamples. In this note, we discuss one such construction, illustrating a potential failure case of an identifiability result presented in "Desiderata for Representation Learning: A Causal Perspective" by Wang & Jordan (2021). The construction is based on the theory of nonlinear independent component analysis. We comment on implications of this and other counterexamples for identifiable representation learning.

Machine learning is the subfield of computer science, that provides computers the ability to automatically learn on their own and improve from their experiences without being explicitly programmed. Though it has been hidden in the recent past but still machine learning has become a basic pillar of IT. We are constantly being surrounded by several ML-based applications like search engines, anti-spam filters, credit card fraud detection system, etc etc. ML is the subset of Artificial Intelligence, which deals with structured and semi-structured data. To understand the concept behind ML, a precise overview of AI is necessary. When AI was coined first in 1955, it's aim was to make machines that are able to perform unique human-based tasks that require intelligence.

Machine learning is embedded in AI and allows machines to perform specific tasks through training. With data annotation, it can learn about pretty much everything. Supervised Learning: The supervised learning learns from a set of labeled data. It is an algorithm that predicts the outcome of new data based on previously known labeled data. Unsupervised Learning: In unsupervised machine learning, training is based on unlabeled data. In this algorithm, you don't know the outcome or the label of the input data.

Machine Learning can be divided into three different types of learning: Unsupervised Learning, Supervised Learning, and Semi-supervised Learning. Unsupervised learning uses information data that is not labeled, that way the machine should work with no guidance according to patterns, similarities, and differences. On the other hand, supervised learning has a presence of a "teacher", who is in charge of training the machine by labeling the data to work with. Next, the machine receives some examples that allow it to produce a correct outcome. But there's a hybrid approach for these types of learning, this Semi-supervised learning works with both labeled and unlabeled data. This method uses a tiny data set of labeled data to train and label the rest of the data with corresponding predictions, finally giving a solution to the problem.

I recently came across the Stanford MedAI Youtube Channel where each week a speaker is invited to give a talk on a topic related to Medical AI, and I would highly recommend you check them out. In this week's talk by Jason Jeong on the applications of Generative Adversarial Networks (GANs) in Medical Imaging, he mentions the paper titled "How to Fool Radiologists with Generative Adversarial Networks?". I was immediately drawn to the clever title and went to check the paper out myself. In this paper published in 2018 by Chuquicusma et al., unsupervised learning with Deep Convolutional Generative Adversarial Networks (DC-GANs) was used to generate realistic-looking images of lung nodules. Two radiologists were then asked to undertake a visual turing test where they were asked to determine whether a lung nodule was fake or real.