In this work, we perform a wide variety of experiments with different deep learning architectures on datasets of limited size. According to our study, we show that model complexity is a critical factor when only a few samples per class are available. Differently from the literature, we show that in some configurations, the state of the art can be improved using low complexity models. For instance, in problems with scarce training samples and without data augmentation, low-complexity convolutional neural networks perform comparably well or better than state-of-the-art architectures. Moreover, we show that even standard data augmentation can boost recognition performance by large margins. This result suggests the development of more complex data generation/augmentation pipelines for cases when data is limited. Finally, we show that dropout, a widely used regularization technique, maintains its role as a good regularizer even when data is scarce. Our findings are empirically validated on the sub-sampled versions of popular CIFAR-10, Fashion-MNIST and, SVHN benchmarks.
Montréal Institute for Learning Algorithms（蒙特利尔学习算法研究所）；Université de Montréal（蒙特利尔大学）；Jagiellonian University（雅盖隆大学）；McGill University（麦吉尔大学）；University of California, Berkeley（加州大学伯克利分校）；Polytechnique Montréal（蒙特利尔理工学院）；University of Bonn（波恩大学）
We examine the role of memorization in deep learning, drawing connections to capacity, generalization, and adversarial robustness. While deep networks are capable of memorizing noise data, our results suggest that they tend to prioritize learning simple patterns first. In our experiments, we expose qualitative differences in gradient-based optimization of deep neural networks (DNNs) on noise vs. real data. We also demonstrate that for appropriately tuned explicit regularization (e.g., dropout) we can degrade DNN training performance on noise datasets without compromising generalization on real…
National Key Laboratory for Novel Software Technology（计算机软件新技术国家重点实验室）；Nanjing University（南京大学）
Deep neural networks need large amounts of labeled data to achieve good performance. In real-world applications, labels are usually collected from non-experts such as crowdsourcing to save cost and thus are noisy. In the past few years, deep learning methods for dealing with noisy labels have been developed, many of which are based on the small-loss criterion. However, there are few theoretical analyses to explain why these methods could learn well from noisy labels. In this paper, we theoretically explain why the widely-used small-loss criterion works. Based on the explanation, we reformalize…
Many decisions involve choosing an uncertain course of actions in deep and wide decision trees, as when we plan to visit an exotic country for vacation. In these cases, exhaustive search for the best sequence of actions is not tractable due to the large number of possibilities and limited time or computational resources available to make the decision. Therefore, planning agents need to balance breadth (exploring many actions at each level of the tree) and depth (exploring many levels in the tree) to allocate optimally their finite search capacity. We provide efficient analytical solutions and …
Practical results have shown that deep learning optimizers using small constant learning rates, hyperparameters close to one, and large batch sizes can find the model parameters of deep neural networks that minimize the loss functions. We first show theoretical evidence that the momentum method (Momentum) and adaptive moment estimation (Adam) perform well in the sense that the upper bound of the theoretical performance measure is small with a small constant learning rate, hyperparameters close to one, and a large batch size. Next, we show that there exists a batch size called the critical batc…
Lead optimization is a key step in drug discovery to produce potent and selective compounds. Historically, in silico screening and structure-based small molecule designing facilitated the processes. Although the recent application of deep learning to drug discovery piloted the possibility of their in silico application lead optimization steps, the real-world application is lacking due to the tool availability. Here, we developed a single user interface application, called Deep2Lead. Our web-based application integrates VAE and DeepPurpose DTI and allows a user to quickly perform a lead optimiz…
Deep Neural Networks (DNNs) are usually over-parameterized, causing excessive memory and interconnection cost on the hardware platform. Existing pruning approaches remove secondary parameters at the end of training to reduce the model size; but without exploiting the intrinsic network property, they still require the full interconnection to prepare the network. Inspired by the observation that brain networks follow the Small-World model, we propose a novel structural pruning scheme, which includes (1) hierarchically trimming the network into a Small-World model before training, (2) training th…
People with type 1 diabetes (T1D) lack the ability to produce the insulin their bodies need. As a result, they must continually make decisions about how much insulin to self-administer to adequately control their blood glucose levels. Longitudinal data streams captured from wearables, like continuous glucose monitors, can help these individuals manage their health, but currently the majority of the decision burden remains on the user. To relieve this burden, researchers are working on closed-loop solutions that combine a continuous glucose monitor and an insulin pump with a control algorithm i…
Two things seem to be indisputable in the contemporary deep learning discourse: 1. The categorical cross-entropy loss after softmax activation is the method of choice for classification. 2. Training a CNN classifier from scratch on small datasets does not work well. In contrast to this, we show that the cosine loss function provides significantly better performance than cross-entropy on datasets with only a handful of samples per class. For example, the accuracy achieved on the CUB-200-2011 dataset without pre-training is by 30% higher than with the cross-entropy loss. Further experiments on o…
We propose a distributed approach to train deep neural networks (DNNs), which has guaranteed convergence theoretically and great scalability empirically: close to 6 times faster on instance of ImageNet data set when run with 6 machines. The proposed scheme is close to optimally scalable in terms of number of machines, and guaranteed to converge to the same optima as the undistributed setting. The convergence and scalability of the distributed setting is shown empirically across different datasets (TIMIT and ImageNet) and machine learning tasks (image classification and phoneme extraction). The…
The skip-connections used in residual networks have become a standard architecture choice in deep learning due to the increased training stability and generalization performance with this architecture, although there has been limited theoretical understanding for this improvement. In this work, we analyze overparameterized deep residual networks trained by gradient descent following random initialization, and demonstrate that (i) the class of networks learned by gradient descent constitutes a small subset of the entire neural network function class, and (ii) this subclass of networks is suffic…
Capital Fund Management（资本基金管理公司）；École Centrale-Supélec（中央理工-高等电力学院）；École Polytechnique（巴黎综合理工学院）
Can deep reinforcement learning algorithms be exploited as solvers for optimal trading strategies? The aim of this work is to test reinforcement learning algorithms on conceptually simple, but mathematically non-trivial, trading environments. The environments are chosen such that an optimal or close-to-optimal trading strategy is known. We study the deep deterministic policy gradient algorithm and show that such a reinforcement learning agent can successfully recover the essential features of the optimal trading strategies and achieve close-to-optimal rewards.
University of Rostock（罗斯托克大学）；Leibniz-Institute for Food Systems Biology（莱布尼茨食品系统生物学研究所）；Technical University of Munich（慕尼黑工业大学）；Technische Universität Dresden（德累斯顿工业大学）
Data is commonly stored in tabular format. Several fields of research are prone to small imbalanced tabular data. Supervised Machine Learning on such data is often difficult due to class imbalance. Synthetic data generation, i.e., oversampling, is a common remedy used to improve classifier performance. State-of-the-art linear interpolation approaches, such as LoRAS and ProWRAS can be used to generate synthetic samples from the convex space of the minority class to improve classifier performance in such cases. Deep generative networks are common deep learning approaches for synthetic sample gen…
We propose a new deep learning approach for medical imaging that copes with the problem of a small training set, the main bottleneck of deep learning, and apply it for classification of healthy and cancer cells acquired by quantitative phase imaging. The proposed method, called transferring of pre-trained generative adversarial network (TOP-GAN), is a hybridization between transfer learning and generative adversarial networks (GANs). Healthy cells and cancer cells of different metastatic potential have been imaged by low-coherence off-axis holography. After the acquisition, the optical path de…
The growth and success of deep learning approaches can be attributed to two major factors: availability of hardware resources and availability of large number of training samples. For problems with large training databases, deep learning models have achieved superlative performances. However, there are a lot of \textit{small sample size or $S^3$} problems for which it is not feasible to collect large training databases. It has been observed that deep learning models do not generalize well on $S^3$ problems and specialized solutions are required. In this paper, we first present a review of deep…
Nara Institute of Science and Technology（奈良先端科学技术大学院大学）
Despite the close relationship between speech perception and production, research in automatic speech recognition (ASR) and text-to-speech synthesis (TTS) has progressed more or less independently without exerting much mutual influence on each other. In human communication, on the other hand, a closed-loop speech chain mechanism with auditory feedback from the speaker's mouth to her ear is crucial. In this paper, we take a step further and develop a closed-loop speech chain model based on deep learning. The sequence-to-sequence model in close-loop architecture allows us to train our model on t…
Now a days, UAVs such as drones are greatly used for various purposes like that of capturing and target detection from ariel imagery etc. Easy access of these small ariel vehicles to public can cause serious security threats. For instance, critical places may be monitored by spies blended in public using drones. Study in hand proposes an improved and efficient Deep Learning based autonomous system which can detect and track very small drones with great precision. The proposed system consists of a custom deep learning model Tiny YOLOv3, one of the flavors of very fast object detection model You…
The current state of cancer therapeutics has been moving away from one-size-fits-all cytotoxic chemotherapy, and towards a more individualized and specific approach involving the targeting of each tumor's genetic vulnerabilities. Different tumors, even of the same type, may be more reliant on certain cellular pathways more than others. With modern advancements in our understanding of cancer genome sequencing, these pathways can be discovered. Investigating each of the millions of possible small molecule inhibitors for each kinase in vitro, however, would be extremely expensive and time consumi…
Centre for Speech Technology Research, The University of Edinburgh（爱丁堡大学语音技术研究中心）
For speech recognition, deep neural networks (DNNs) have significantly improved the recognition accuracy in most of benchmark datasets and application domains. However, compared to the conventional Gaussian mixture models, DNN-based acoustic models usually have much larger number of model parameters, making it challenging for their applications in resource constrained platforms, e.g., mobile devices. In this paper, we study the application of the recently proposed highway network to train small-footprint DNNs, which are {\it thinner} and {\it deeper}, and have significantly smaller number of m…
The University of Western Australia（西澳大学）；The University of Melbourne（墨尔本大学）；Defence Science and Technology Group（国防科学技术集团）
Small object detection (SOD) in optical images and videos is a challenging problem that even state-of-the-art generic object detection methods fail to accurately localize and identify such objects. Typically, small objects appear in real-world due to large camera-object distance. Because small objects occupy only a small area in the input image (e.g., less than 10%), the information extracted from such a small area is not always rich enough to support decision making. Multidisciplinary strategies are being developed by researchers working at the interface of deep learning and computer vision t…
Borealis AI；University of Waterloo（滑铁卢大学）；University of Alberta（阿尔伯塔大学）；McGill University（麦吉尔大学）；Mila（米拉人工智能研究所）
It is a common belief that training deep transformers from scratch requires large datasets. Consequently, for small datasets, people usually use shallow and simple additional layers on top of pre-trained models during fine-tuning. This work shows that this does not always need to be the case: with proper initialization and optimization, the benefits of very deep transformers can carry over to challenging tasks with small datasets, including Text-to-SQL semantic parsing and logical reading comprehension. In particular, we successfully train $48$ layers of transformers, comprising $24$ fine-tune…
