As an agent moves through the world, the apparent motion of scene elements is (usually) inversely proportional to their depth. It is natural for a learning agent to associate image patterns with the magnitude of their displacement over time: as the agent moves, faraway mountains don't move much; nearby trees move a lot. This natural relationship between the appearance of objects and their motion is a rich source of information about the world. In this work, we start by training a deep network, using fully automatic supervision, to predict relative scene depth from single images. The relative depth training images are automatically derived from simple videos of cars moving through a scene, using recent motion segmentation techniques, and no human-provided labels. This proxy task of predicting relative depth from a single image induces features in the network that result in large improvements in a set of downstream tasks including semantic segmentation, joint road segmentation and car detection, and monocular (absolute) depth estimation, over a network trained from scratch. The improvement on the semantic segmentation task is greater than those produced by any other automatically supervised methods. Moreover, for monocular depth estimation, our unsupervised pre-training method even outperforms supervised pre-training with ImageNet. In addition, we demonstrate benefits from learning to predict (unsupervised) relative depth in the specific videos associated with various downstream tasks. We adapt to the specific scenes in those tasks in an unsupervised manner to improve performance. In summary, for semantic segmentation, we present state-of-the-art results among methods that do not use supervised pre-training, and we even exceed the performance of supervised ImageNet pre-trained models for monocular depth estimation, achieving results that are comparable with state-of-the-art methods.

Click to Read Paper
The recent explosive growth in convolutional neural network (CNN) research has produced a variety of new architectures for deep learning. One intriguing new architecture is the bilinear CNN (B-CNN), which has shown dramatic performance gains on certain fine-grained recognition problems [15]. We apply this new CNN to the challenging new face recognition benchmark, the IARPA Janus Benchmark A (IJB-A) [12]. It features faces from a large number of identities in challenging real-world conditions. Because the face images were not identified automatically using a computerized face detection system, it does not have the bias inherent in such a database. We demonstrate the performance of the B-CNN model beginning from an AlexNet-style network pre-trained on ImageNet. We then show results for fine-tuning using a moderate-sized and public external database, FaceScrub [17]. We also present results with additional fine-tuning on the limited training data provided by the protocol. In each case, the fine-tuned bilinear model shows substantial improvements over the standard CNN. Finally, we demonstrate how a standard CNN pre-trained on a large face database, the recently released VGG-Face model [20], can be converted into a B-CNN without any additional feature training. This B-CNN improves upon the CNN performance on the IJB-A benchmark, achieving 89.5% rank-1 recall.

* Published version at WACV 2016
Click to Read Paper
The Intel Collaborative Research Institute for Computational Intelligence (ICRI-CI) has been heavily supporting Machine Learning and Deep Learning research from its foundation in 2012. We have asked six leading ICRI-CI Deep Learning researchers to address the challenge of "Why & When Deep Learning works", with the goal of looking inside Deep Learning, providing insights on how deep networks function, and uncovering key observations on their expressiveness, limitations, and potential. The output of this challenge resulted in five papers that address different facets of deep learning. These different facets include a high-level understating of why and when deep networks work (and do not work), the impact of geometry on the expressiveness of deep networks, and making deep networks interpretable.

* This paper is the preface part of the "Why & When Deep Learning works looking inside Deep Learning" ICRI-CI paper bundle
Click to Read Paper
Few-shot learning remains challenging for meta-learning that learns a learning algorithm (meta-learner) from many related tasks. In this work, we argue that this is due to the lack of a good representation for meta-learning, and propose deep meta-learning to integrate the representation power of deep learning into meta-learning. The framework is composed of three modules, a concept generator, a meta-learner, and a concept discriminator, which are learned jointly. The concept generator, e.g. a deep residual net, extracts a representation for each instance that captures its high-level concept, on which the meta-learner performs few-shot learning, and the concept discriminator recognizes the concepts. By learning to learn in the concept space rather than in the complicated instance space, deep meta-learning can substantially improve vanilla meta-learning, which is demonstrated on various few-shot image recognition problems. For example, on 5-way-1-shot image recognition on CIFAR-100 and CUB-200, it improves Matching Nets from 50.53% and 56.53% to 58.18% and 63.47%, improves MAML from 49.28% and 50.45% to 56.65% and 64.63%, and improves Meta-SGD from 53.83% and 53.34% to 61.62% and 66.95%, respectively.

Click to Read Paper
Deep learning has achieved a great success in many areas, from computer vision to natural language processing, to game playing, and much more. Yet, what deep learning is really doing is still an open question. There are a lot of works in this direction. For example, [5] tried to explain deep learning by group renormalization, and [6] tried to explain deep learning from the view of functional approximation. In order to address this very crucial question, here we see deep learning from perspective of mechanical learning and learning machine (see [1], [2]). From this particular angle, we can see deep learning much better and answer with confidence: What deep learning is really doing? why it works well, how it works, and how much data is necessary for learning. We also will discuss advantages and disadvantages of deep learning at the end of this work.

Click to Read Paper
Why and how that deep learning works well on different tasks remains a mystery from a theoretical perspective. In this paper we draw a geometric picture of the deep learning system by finding its analogies with two existing geometric structures, the geometry of quantum computations and the geometry of the diffeomorphic template matching. In this framework, we give the geometric structures of different deep learning systems including convolutional neural networks, residual networks, recursive neural networks, recurrent neural networks and the equilibrium prapagation framework. We can also analysis the relationship between the geometrical structures and their performance of different networks in an algorithmic level so that the geometric framework may guide the design of the structures and algorithms of deep learning systems.

* 16 pages, 13 figures
Click to Read Paper
Deep Neural Networks (DNNs) are typically trained by backpropagation in a batch learning setting, which requires the entire training data to be made available prior to the learning task. This is not scalable for many real-world scenarios where new data arrives sequentially in a stream form. We aim to address an open challenge of "Online Deep Learning" (ODL) for learning DNNs on the fly in an online setting. Unlike traditional online learning that often optimizes some convex objective function with respect to a shallow model (e.g., a linear/kernel-based hypothesis), ODL is significantly more challenging since the optimization of the DNN objective function is non-convex, and regular backpropagation does not work well in practice, especially for online learning settings. In this paper, we present a new online deep learning framework that attempts to tackle the challenges by learning DNN models of adaptive depth from a sequence of training data in an online learning setting. In particular, we propose a novel Hedge Backpropagation (HBP) method for online updating the parameters of DNN effectively, and validate the efficacy of our method on large-scale data sets, including both stationary and concept drifting scenarios.

Click to Read Paper
The great success of deep learning shows that its technology contains profound truth, and understanding its internal mechanism not only has important implications for the development of its technology and effective application in various fields, but also provides meaningful insights into the understanding of human brain mechanism. At present, most of the theoretical research on deep learning is based on mathematics. This dissertation proposes that the neural network of deep learning is a physical system, examines deep learning from three different perspectives: microscopic, macroscopic, and physical world views, answers multiple theoretical puzzles in deep learning by using physics principles. For example, from the perspective of quantum mechanics and statistical physics, this dissertation presents the calculation methods for convolution calculation, pooling, normalization, and Restricted Boltzmann Machine, as well as the selection of cost functions, explains why deep learning must be deep, what characteristics are learned in deep learning, why Convolutional Neural Networks do not have to be trained layer by layer, and the limitations of deep learning, etc., and proposes the theoretical direction and basis for the further development of deep learning now and in the future. The brilliance of physics flashes in deep learning, we try to establish the deep learning technology based on the scientific theory of physics.

Click to Read Paper
Deep reinforcement learning is revolutionizing the artificial intelligence field. Currently, it serves as a good starting point for constructing intelligent autonomous systems which offer a better knowledge of the visual world. It is possible to scale deep reinforcement learning with the use of deep learning and do amazing tasks such as use of pixels in playing video games. In this paper, key concepts of deep reinforcement learning including reward function, differences between reinforcement learning and supervised learning and models for implementation of reinforcement are discussed. Key challenges related to the implementation of reinforcement learning in conversational AI domain are identified as well as discussed in detail. Various conversational models which are based on deep reinforcement learning (as well as deep learning) are also discussed. In summary, this paper discusses key aspects of deep reinforcement learning which are crucial for designing an efficient conversational AI.

* SCAI'17-Search-Oriented Conversational AI (@ICTIR)
Click to Read Paper
Deep Learning is one of the newest trends in Machine Learning and Artificial Intelligence research. It is also one of the most popular scientific research trends now-a-days. Deep learning methods have brought revolutionary advances in computer vision and machine learning. Every now and then, new and new deep learning techniques are being born, outperforming state-of-the-art machine learning and even existing deep learning techniques. In recent years, the world has seen many major breakthroughs in this field. Since deep learning is evolving at a huge speed, its kind of hard to keep track of the regular advances especially for new researchers. In this paper, we are going to briefly discuss about recent advances in Deep Learning for past few years.

* 31 pages including bibliography
Click to Read Paper
Deep Learning has recently become hugely popular in machine learning, providing significant improvements in classification accuracy in the presence of highly-structured and large databases. Researchers have also considered privacy implications of deep learning. Models are typically trained in a centralized manner with all the data being processed by the same training algorithm. If the data is a collection of users' private data, including habits, personal pictures, geographical positions, interests, and more, the centralized server will have access to sensitive information that could potentially be mishandled. To tackle this problem, collaborative deep learning models have recently been proposed where parties locally train their deep learning structures and only share a subset of the parameters in the attempt to keep their respective training sets private. Parameters can also be obfuscated via differential privacy (DP) to make information extraction even more challenging, as proposed by Shokri and Shmatikov at CCS'15. Unfortunately, we show that any privacy-preserving collaborative deep learning is susceptible to a powerful attack that we devise in this paper. In particular, we show that a distributed, federated, or decentralized deep learning approach is fundamentally broken and does not protect the training sets of honest participants. The attack we developed exploits the real-time nature of the learning process that allows the adversary to train a Generative Adversarial Network (GAN) that generates prototypical samples of the targeted training set that was meant to be private (the samples generated by the GAN are intended to come from the same distribution as the training data). Interestingly, we show that record-level DP applied to the shared parameters of the model, as suggested in previous work, is ineffective (i.e., record-level DP is not designed to address our attack).

* ACM CCS'17, 16 pages, 18 figures
Click to Read Paper
Deep Learning is a very powerful machine learning model. Deep Learning trains a large number of parameters for multiple layers and is very slow when data is in large scale and the architecture size is large. Inspired from the shrinking technique used in accelerating computation of Support Vector Machines (SVM) algorithm and screening technique used in LASSO, we propose a shrinking Deep Learning with recall (sDLr) approach to speed up deep learning computation. We experiment shrinking Deep Learning with recall (sDLr) using Deep Neural Network (DNN), Deep Belief Network (DBN) and Convolution Neural Network (CNN) on 4 data sets. Results show that the speedup using shrinking Deep Learning with recall (sDLr) can reach more than 2.0 while still giving competitive classification performance.

* The 22nd IEEE International Conference on Parallel and Distributed Systems (ICPADS 2016)
Click to Read Paper
This paper presents a generic Bayesian framework that enables any deep learning model to actively learn from targeted crowds. Our framework inherits from recent advances in Bayesian deep learning, and extends existing work by considering the targeted crowdsourcing approach, where multiple annotators with unknown expertise contribute an uncontrolled amount (often limited) of annotations. Our framework leverages the low-rank structure in annotations to learn individual annotator expertise, which then helps to infer the true labels from noisy and sparse annotations. It provides a unified Bayesian model to simultaneously infer the true labels and train the deep learning model in order to reach an optimal learning efficacy. Finally, our framework exploits the uncertainty of the deep learning model during prediction as well as the annotators' estimated expertise to minimize the number of required annotations and annotators for optimally training the deep learning model. We evaluate the effectiveness of our framework for intent classification in Alexa (Amazon's personal assistant), using both synthetic and real-world datasets. Experiments show that our framework can accurately learn annotator expertise, infer true labels, and effectively reduce the amount of annotations in model training as compared to state-of-the-art approaches. We further discuss the potential of our proposed framework in bridging machine learning and crowdsourcing towards improved human-in-the-loop systems.

Click to Read Paper
Researchers have proposed that deep learning, which is providing important progress in a wide range of high complexity tasks, might inspire new insights into learning in the brain. However, the methods used for deep learning by artificial neural networks are biologically unrealistic and would need to be replaced by biologically realistic counterparts. Previous biologically plausible reinforcement learning rules, like AGREL and AuGMEnT, showed promising results but focused on shallow networks with three layers. Will these learning rules also generalize to networks with more layers and can they handle tasks of higher complexity? Here, we demonstrate that these learning schemes indeed generalize to deep networks, if we include an attention network that propagates information about the selected action to lower network levels. The resulting learning rule, called Q-AGREL, is equivalent to a particular form of error-backpropagation that trains one output unit at any one time. To demonstrate the utility of the learning scheme for larger problems, we trained networks with two hidden layers on the MNIST dataset, a standard and interesting Machine Learning task. Our results demonstrate that the capability of Q-AGREL is comparable to that of error backpropagation, although the learning rate is 1.5-2 times slower because the network has to learn by trial-and-error and updates the action value of only one output unit at a time. Our results provide new insights into how deep learning can be implemented in the brain.

Click to Read Paper
Deep learning is very effective at jointly learning feature representations and classification models, especially when dealing with high dimensional input patterns. Probabilistic logic reasoning, on the other hand, is capable to take consistent and robust decisions in complex environments. The integration of deep learning and logic reasoning is still an open-research problem and it is considered to be the key for the development of real intelligent agents. This paper presents Deep Logic Models, which are deep graphical models integrating deep learning and logic reasoning both for learning and inference. Deep Logic Models create an end-to-end differentiable architecture, where deep learners are embedded into a network implementing a continuous relaxation of the logic knowledge. The learning process allows to jointly learn the weights of the deep learners and the meta-parameters controlling the high-level reasoning. The experimental results show that the proposed methodology overtakes the limitations of the other approaches that have been proposed to bridge deep learning and reasoning.

Click to Read Paper
How to understand deep learning systems remains an open problem. In this paper we propose that the answer may lie in the geometrization of deep networks. Geometrization is a bridge to connect physics, geometry, deep network and quantum computation and this may result in a new scheme to reveal the rule of the physical world. By comparing the geometry of image matching and deep networks, we show that geometrization of deep networks can be used to understand existing deep learning systems and it may also help to solve the interpretability problem of deep learning systems.

* 9 pages, draft version
Click to Read Paper
Recently, deep learning techniques have enjoyed success in various multimedia applications, such as image classification and multi-modal data analysis. Large deep learning models are developed for learning rich representations of complex data. There are two challenges to overcome before deep learning can be widely adopted in multimedia and other applications. One is usability, namely the implementation of different models and training algorithms must be done by non-experts without much effort especially when the model is large and complex. The other is scalability, that is the deep learning system must be able to provision for a huge demand of computing resources for training large models with massive datasets. To address these two challenges, in this paper, we design a distributed deep learning platform called SINGA which has an intuitive programming model based on the common layer abstraction of deep learning models. Good scalability is achieved through flexible distributed training architecture and specific optimization techniques. SINGA runs on GPUs as well as on CPUs, and we show that it outperforms many other state-of-the-art deep learning systems. Our experience with developing and training deep learning models for real-life multimedia applications in SINGA shows that the platform is both usable and scalable.

* submitted to TOMM (under review)
Click to Read Paper
Deep learning has made significant breakthroughs in various fields of artificial intelligence. Advantages of deep learning include the ability to capture highly complicated features, weak involvement of human engineering, etc. However, it is still virtually impossible to use deep learning to analyze programs since deep architectures cannot be trained effectively with pure back propagation. In this pioneering paper, we propose the "coding criterion" to build program vector representations, which are the premise of deep learning for program analysis. Our representation learning approach directly makes deep learning a reality in this new field. We evaluate the learned vector representations both qualitatively and quantitatively. We conclude, based on the experiments, the coding criterion is successful in building program representations. To evaluate whether deep learning is beneficial for program analysis, we feed the representations to deep neural networks, and achieve higher accuracy in the program classification task than "shallow" methods, such as logistic regression and the support vector machine. This result confirms the feasibility of deep learning to analyze programs. It also gives primary evidence of its success in this new field. We believe deep learning will become an outstanding technique for program analysis in the near future.

* This paper was submitted to ICSE'14
Click to Read Paper
In this work we propose a new deep learning tool called deep dictionary learning. Multi-level dictionaries are learnt in a greedy fashion, one layer at a time. This requires solving a simple (shallow) dictionary learning problem, the solution to this is well known. We apply the proposed technique on some benchmark deep learning datasets. We compare our results with other deep learning tools like stacked autoencoder and deep belief network; and state of the art supervised dictionary learning tools like discriminative KSVD and label consistent KSVD. Our method yields better results than all.

Click to Read Paper
In the era of big data, transformation of biomedical big data into valuable knowledge has been one of the most important challenges in bioinformatics. Deep learning has advanced rapidly since the early 2000s and now demonstrates state-of-the-art performance in various fields. Accordingly, application of deep learning in bioinformatics to gain insight from data has been emphasized in both academia and industry. Here, we review deep learning in bioinformatics, presenting examples of current research. To provide a useful and comprehensive perspective, we categorize research both by the bioinformatics domain (i.e., omics, biomedical imaging, biomedical signal processing) and deep learning architecture (i.e., deep neural networks, convolutional neural networks, recurrent neural networks, emergent architectures) and present brief descriptions of each study. Additionally, we discuss theoretical and practical issues of deep learning in bioinformatics and suggest future research directions. We believe that this review will provide valuable insights and serve as a starting point for researchers to apply deep learning approaches in their bioinformatics studies.

* Accepted for Briefings in Bioinformatics (18-Jun-2016)
Click to Read Paper