It was clear when Ouster started developing the OS-1 three years ago that deep learning research for cameras was outpacing lidar research. Lidar data has incredible benefits -- rich spatial information and lighting agnostic sensing to name a couple -- but it lacks the raw resolution and efficient array structure of camera images, and 3D point clouds are still more difficult to encode in a neural net or process with hardware acceleration. With the tradeoffs between both sensing modalities in mind, we set out to bring the best aspects of lidars and cameras together in a single device from the very beginning. Today we're releasing a firmware upgrade and update to our open source driver that deliver on that goal. Our OS-1 lidar now outputs fixed resolution depth images, signal images, and ambient images in real time, all without a camera.

Neural networks trained with backpropagation, the standard algorithm of deep learning which uses weight transport, are easily fooled by existing gradient-based adversarial attacks. This class of attacks are based on certain small perturbations of the inputs to make networks misclassify them. We show that less biologically implausible deep neural networks trained with feedback alignment, which do not use weight transport, can be harder to fool, providing actual robustness. Tested on MNIST, deep neural networks trained without weight transport (1) have an adversarial accuracy of 98% compared to 0.03% for neural networks trained with backpropagation and (2) generate non-transferable adversarial examples. However, this gap decreases on CIFAR-10 but still significant particularly for small perturbation magnitude less than 1/2.

--In this paper, a method of prediction on continuous time series variables from the production or flow - an LSTM algorithm based on multivariate tuning - is proposed. The algorithm improves the traditional LSTM algorithm and converts the time series data into supervised learning sequences regarding industrial data's features. The main innovation of this paper consists in introducing the concepts of periodic measurement and time window in the industrial prediction problem, especially considering industrial data with time series characteristics. Experiments using real-world datasets show that the prediction accuracy is improved, 54.05% higher than that of traditional LSTM algorithm. In industry, with the high-speed functioning of the enterprise product line, data are generated continuously. Malfunctions and abnormality often take place, which incur a great deal of money and resources, and even advanced equipment cannot avoid these problems[1]. Industrial companies have to pay a lot to maintain and ensure the normal operation of the manufacturing process. According to the statistics, the maintenance costs of all kinds of industrial enterprises account for about 15%-70% of total production costs[2]. Flow prediction is motivated by such industrial conundrums faced by many factories. Implementing flow prediction to forecast the output of the machine and to detect the problems in time via prediction, not only can production increase, but also a large number of workforce and resources for troubleshooting can be saved.

Deep neural networks achieve state-of-the-art results on several tasks while increasing in complexity. It has been shown that neural networks can be pruned during training by imposing sparsity inducing regularizers. In this paper, we investigate two techniques for group-wise pruning during training in order to improve network efficiency. We propose a gating factor after every convolutional layer to induce channel level sparsity, encouraging insignificant channels to become exactly zero. Further, we introduce and analyse a bounded variant of the L1 regularizer, which interpolates between L1 and L0-norms to retain performance of the network at higher pruning rates. To underline effectiveness of the proposed methods,we show that the number of parameters of ResNet-164, DenseNet-40 and MobileNetV2 can be reduced down by 30%, 69% and 75% on CIFAR100 respectively without a significant drop in accuracy. We achieve state-of-the-art pruning results for ResNet-50 with higher accuracy on ImageNet. Furthermore, we show that the light weight MobileNetV2 can further be compressed on ImageNet without a significant drop in performance.

Path-planning algorithms are an important part of a wide variety of robotic applications, such as mobile robot navigation and robot arm manipulation. However, in large search spaces in which local traps may exist, it remains challenging to reliably find a path while satisfying real-time constraints. Efforts to speed up the path search have led to the development of many practical path-planning algorithms. These algorithms often define a search heuristic to guide the search towards the goal. The heuristics should be carefully designed for each specific problem to ensure reliability in the various situations encountered in the problem. However, it is often difficult for humans to craft such robust heuristics, and the search performance often degrades under conditions that violate the heuristic assumption. Rather than manually designing the heuristics, in this work, we propose a learning approach to acquire these search heuristics. Our method represents the environment containing the obstacles as an image, and this image is fed into fully convolutional neural networks to produce a search heuristic image where every pixel represents a heuristic value (cost-to-go value to a goal) in the form of a vertex of a search graph. Training the heuristic is performed using previously collected planning results. Our preliminary experiments (2D grid world navigation experiments) demonstrate significant reduction in the search costs relative to a hand-designed heuristic.

The majority of contemporary object-tracking approaches used in autonomous vehicles do not model interactions between objects. This contrasts with the fact that objects' paths are not independent: a cyclist might abruptly deviate from a previously planned trajectory in order to avoid colliding with a car. Building upon HART, a neural, class-agnostic single-object tracker, we introduce a multi-object tracking method MOHART capable of relational reasoning. Importantly, the entire system, including the understanding of interactions and relations between objects, is class-agnostic and learned simultaneously in an end-to-end fashion. We find that the addition of relational-reasoning capabilities to HART leads to consistent performance gains in tracking as well as future trajectory prediction on several real-world datasets (MOTChallenge, UA-DETRAC, and Stanford Drone dataset), particularly in the presence of ego-motion, occlusions, crowded scenes, and faulty sensor inputs. Finally, based on controlled simulations, we propose that a comparison of MOHART and HART may be used as a novel way to measure the degree to which the objects in a video depend upon each other as they move together through time.

Explaining the prediction of deep neural networks (DNNs) and semantic image compression are two active research areas of deep learning with a numerous of applications in decision-critical systems, such as surveillance cameras, drones and self-driving cars, where interpretable decision is critical and storage/network bandwidth is limited. In this paper, we propose a novel end-to-end Neural Image Compression and Explanation (NICE) framework that learns to (1) explain the prediction of convolutional neural networks (CNNs), and (2) subsequently compress the input images for efficient storage or transmission. Specifically, NICE generates a sparse mask over an input image by attaching a stochastic binary gate to each pixel of the image, whose parameters are learned through the interaction with the CNN classifier to be explained. The generated mask is able to capture the saliency of each pixel measured by its influence to the final prediction of CNN; it can also be used to produce a mixed-resolution image, where important pixels maintain their original high resolution and insignificant background pixels are subsampled to a low resolution. The produced images achieve a high compression rate (e.g., about 0.6x of original image file size), while retaining a similar classification accuracy. Extensive experiments across multiple image classification benchmarks demonstrate the superior performance of NICE compared to the state-of-the-art methods in terms of explanation quality and image compression rate.

Variational inference is becoming more and more popular for approximating intractable posterior distributions in Bayesian statistics and machine learning. Meanwhile, a few recent works have provided theoretical justification and new insights on deep neural networks for estimating smooth functions in usual settings such as nonparametric regression. In this paper, we show that variational inference for sparse deep learning retains the same generalization properties than exact Bayesian inference. In particular, we highlight the connection between estimation and approximation theories via the classical bias-variance trade-off and show that it leads to near-minimax rates of convergence for H\"older smooth functions. Additionally, we show that the model selection framework over the neural network architecture via ELBO maximization does not overfit and adaptively achieves the optimal rate of convergence.

In recent years there has been a growing interest in image generation through deep learning. While an important part of the evaluation of the generated images usually involves visual inspection, the inclusion of human perception as a factor in the training process is often overlooked. In this paper we propose an alternative perceptual regulariser for image-to-image translation using conditional generative adversarial networks (cGANs). To do so automatically (avoiding visual inspection), we use the Normalised Laplacian Pyramid Distance (NLPD) to measure the perceptual similarity between the generated image and the original image. The NLPD is based on the principle of normalising the value of coefficients with respect to a local estimate of mean energy at different scales and has already been successfully tested in different experiments involving human perception. We compare this regulariser with the originally proposed L1 distance and note that when using NLPD the generated images contain more realistic values for both local and global contrast. We found that using NLPD as a regulariser improves image segmentation accuracy on generated images as well as improving two no-reference image quality metrics.

Most recent semi-supervised deep learning (deep SSL) methods used a similar paradigm: use network predictions to update pseudo-labels and use pseudo-labels to update network parameters iteratively. However, they lack theoretical support and can not explain why predictions are good candidates for pseudo-labels. In this paper, we propose a principled end-to-end framework named deep decipher (D2) for SSL. With the D2 framework, we prove that pseudo-labels are related to network predictions by an exponential link function, which gives a theoretical support for using predictions as pseudo-labels. Furthermore, we demonstrate that updating pseudo-labels by network predictions will make them uncertain. To mitigate this problem, we propose a training strategy called repetitive reprediction (R2). Finally, the proposed R2-D2 method is tested on the large-scale ImageNet dataset and outperforms state-of-the-art methods by $5\%$.