Convolutional Neural Networks (CNN) have shown impressive performance in computer vision, natural language processing, and many other applications, but they exhibit high computations and substantial memory requirements. To address these limitations, especially in resource-constrained devices, the use of cloud computing for CNNs is becoming more popular. This comes with privacy and latency concerns that have motivated the designers to develop embedded hardware accelerators for CNNs. However, designing a specialized accelerator increases the time-to-market and cost of production. Therefore, to reduce the time-to-market and access to state-of-the-art techniques, CNN hardware mapping and deployment on embedded accelerators are often outsourced to untrusted third parties, which is going to be more prevalent in futuristic artificial intelligence of things (AIoT) systems. These AIoT systems anticipate horizontal collaboration among different resource-constrained AIoT node devices, where CNN layers are partitioned and these devices collaboratively compute complex CNN tasks Therefore, there is a dire need to explore this attack surface for designing secure embedded hardware accelerators for CNNs. Towards this goal, in this paper, we exploited this attack surface to propose an HT-based attack called FeSHI. This attack exploits the statistical distribution i.e., Gaussian distribution, of the layer-by-layer feature maps of the CNN to design two triggers for stealthy HT with a very low probability of triggering. To illustrate the effectiveness of the proposed attack, we deployed the LeNet and LeNet-3D on PYNQ to classify the MNIST and CIFAR-10 datasets, respectively, and tested FeSHI. The experimental results show that FeSHI utilizes up to 2% extra LUTs, and the overall resource overhead is less than 1% compared to the original designs

Differentiable neural architecture search (NAS) has attracted significant attention in recent years due to its ability to quickly discover promising architectures of deep neural networks even in very large search spaces. Despite its success, DARTS lacks robustness in certain cases, e.g. it may degenerate to trivial architectures with excessive parametric-free operations such as skip connection or random noise, leading to inferior performance. In particular, operation selection based on the magnitude of architectural parameters was recently proven to be fundamentally wrong showcasing the need to rethink this aspect. On the other hand, zero-cost proxies have been recently studied in the context of sample-based NAS showing promising results -- speeding up the search process drastically in some cases but also failing on some of the large search spaces typical for differentiable NAS. In this work we propose a novel operation selection paradigm in the context of differentiable NAS which utilises zero-cost proxies. Our perturbation-based zero-cost operation selection (Zero-Cost-PT) improves searching time and, in many cases, accuracy compared to the best available differentiable architecture search, regardless of the search space size. Specifically, we are able to find comparable architectures to DARTS-PT on the DARTS CNN search space while being over 40x faster (total searching time 25 minutes on a single GPU).

For certain infinitely-wide neural networks, the neural tangent kernel (NTK) theory fully characterizes generalization. However, for the networks used in practice, the empirical NTK represents only a rough first-order approximation of these architectures. Still, a growing body of work keeps leveraging this approximation to successfully analyze important deep learning phenomena and derive algorithms for new applications. In our work, we provide strong empirical evidence to determine the practical validity of such approximation by conducting a systematic comparison of the behaviour of different neural networks and their linear approximations on different tasks. We show that the linear approximations can indeed rank the learning complexity of certain tasks for neural networks, albeit with important nuances. Specifically, we discover that, in contrast to what was previously observed, neural networks do not always perform better than their kernel approximations, and reveal that their performance gap heavily depends on architecture, number of samples and training task. In fact, we show that during training, deep networks increase the alignment of their empirical NTK with the target task, which explains why linear approximations at the end of training can better explain the dynamics of deep networks. Overall, our work provides concrete examples of novel deep learning phenomena which can inspire future theoretical research, as well as provides a new perspective on the use of the NTK approximation in deep learning.

Designing an effective loss function plays a crucial role in training deep recommender systems. Most existing works often leverage a predefined and fixed loss function that could lead to suboptimal recommendation quality and training efficiency. Some recent efforts rely on exhaustively or manually searched weights to fuse a group of candidate loss functions, which is exceptionally costly in computation and time. They also neglect the various convergence behaviors of different data examples. In this work, we propose an AutoLoss framework that can automatically and adaptively search for the appropriate loss function from a set of candidates. To be specific, we develop a novel controller network, which can dynamically adjust the loss probabilities in a differentiable manner. Unlike existing algorithms, the proposed controller can adaptively generate the loss probabilities for different data examples according to their varied convergence behaviors. Such design improves the model's generalizability and transferability between deep recommender systems and datasets. We evaluate the proposed framework on two benchmark datasets. The results show that AutoLoss outperforms representative baselines. Further experiments have been conducted to deepen our understandings of AutoLoss, including its transferability, components and training efficiency.

Large-scale models for learning fixed-dimensional cross-lingual sentence representations like LASER (Artetxe and Schwenk, 2019b) lead to significant improvement in performance on downstream tasks. However, further increases and modifications based on such large-scale models are usually impractical due to memory limitations. In this work, we introduce a lightweight dual-transformer architecture with just 2 layers for generating memory-efficient cross-lingual sentence representations. We explore different training tasks and observe that current cross-lingual training tasks leave a lot to be desired for this shallow architecture. To ameliorate this, we propose a novel cross-lingual language model, which combines the existing single-word masked language model with the newly proposed cross-lingual token-level reconstruction task. We further augment the training task by the introduction of two computationally-lite sentence-level contrastive learning tasks to enhance the alignment of cross-lingual sentence representation space, which compensates for the learning bottleneck of the lightweight transformer for generative tasks. Our comparisons with competing models on cross-lingual sentence retrieval and multilingual document classification confirm the effectiveness of the newly proposed training tasks for a shallow model.

How can neural networks trained by contrastive learning extract features from the unlabeled data? Why does contrastive learning usually need much stronger data augmentations than supervised learning to ensure good representations? These questions involve both the optimization and statistical aspects of deep learning, but can hardly be answered by analyzing supervised learning, where the target functions are the highest pursuit. Indeed, in self-supervised learning, it is inevitable to relate to the optimization/generalization of neural networks to how they can encode the latent structures in the data, which we refer to as the feature learning process. In this work, we formally study how contrastive learning learns the feature representations for neural networks by analyzing its feature learning process. We consider the case where our data are comprised of two types of features: the more semantically aligned sparse features which we want to learn from, and the other dense features we want to avoid. Theoretically, we prove that contrastive learning using $\mathbf{ReLU}$ networks provably learns the desired sparse features if proper augmentations are adopted. We present an underlying principle called $\textbf{feature decoupling}$ to explain the effects of augmentations, where we theoretically characterize how augmentations can reduce the correlations of dense features between positive samples while keeping the correlations of sparse features intact, thereby forcing the neural networks to learn from the self-supervision of sparse features. Empirically, we verified that the feature decoupling principle matches the underlying mechanism of contrastive learning in practice.

Many significant studies have been published in recent past that advocate usage of machine learning (ML) and deep learning models for COVID-19 …

I'm a huge fan of Lex Fridman and the awesome content he produces to promote ideas and advances in different sciences. In this regard, I want to share some of the concepts that blew my mind when I first heard them in his Podcasts. He is the President and Chief Scientific Officer of the Allen Institute for Brain Science in Seattle. From 1986 until 2013, he was a professor at CalTech. Cited more than 105,000 times.

As the name suggests Neural Networks can be defined as a set of algorithms that are specifically designed to act like a human brain. In a technological manner, it can be referred to as a computational model that has a network architecture. To get a general idea here I have listed some of its components.

It is not the best of times for self-driving car startups. The past year has seen large tech companies acquire startups that were running out of cash and ride-hailing companies shutter costly self-driving car projects with no prospect of becoming production-ready anytime soon. Yet, in the midst of this downturn, Waabi, a Toronto-based self-driving car startup, has just come out of stealth with an insane amount of $83.5 million in a Series A funding round led by Khosla Ventures, with additional participation from Uber, 8VC, Radical Ventures, OMERS Ventures, BDC, and Aurora Innovation. The company's financial backers also include Geoffrey Hinton, Fei-Fei Li, Peter Abbeel, and Sanja Fidler, artificial intelligence scientists with great influence in the academia and applied AI community. What makes Waabi qualified for such support?