To implement deep neural networks on small embedded devices, conventional techniques use channel pruning looking considering manual compression rate per layer to reduce parameters. Besides it is difficult to consider the relationships between layers and it takes a lot of time for deeper models. For addressing these issues, we propose a new channel pruning technique based on attention that can evaluate the importance of channels. We improved the method with the criterion to allow the automatic channel selection using a single compression rate for the entire model. Experimental results showed that a parameter reduction of 90.8% and FLOPs reduction of 79.4% was achieved with an accuracy degradation of around 1% for the compressed ResNet-50 model on the CIFAR-10 benchmark.

This paper focuses on the problem of learning 6-DOF grasping with a parallel jaw gripper in simulation. We propose the notion of a geometry-aware representation in grasping based on the assumption that knowledge of 3D geometry is at the heart of interaction. Our key idea is constraining and regularizing grasping interaction learning through 3D geometry prediction. Specifically, we formulate the learning of deep geometry-aware grasping model in two steps: First, we learn to build mental geometry-aware representation by reconstructing the scene (i.e., 3D occupancy grid) from RGBD input via generative 3D shape modeling. Second, we learn to predict grasping outcome with its internal geometry-aware representation. The learned outcome prediction model is used to sequentially propose grasping solutions via analysis-by-synthesis optimization. Our contributions are fourfold: (1) To best of our knowledge, we are presenting for the first time a method to learn a 6-DOF grasping net from RGBD input; (2) We build a grasping dataset from demonstrations in virtual reality with rich sensory and interaction annotations. This dataset includes 101 everyday objects spread across 7 categories, additionally, we propose a data augmentation strategy for effective learning; (3) We demonstrate that the learned geometry-aware representation leads to about 10 percent relative performance improvement over the baseline CNN on grasping objects from our dataset. (4) We further demonstrate that the model generalizes to novel viewpoints and object instances.

Fast and efficient motion planning algorithms are crucial for many state-of-the-art robotics applications such as self-driving cars. Existing motion planning methods such as RRT*, A*, and D*, become ineffective as their computational complexity increases exponentially with the dimensionality of the motion planning problem. To address this issue, we present a neural network-based novel planning algorithm which generates end-to-end collision-free paths irrespective of the obstacles' geometry. The proposed method, called MPNet (Motion Planning Network), comprises of a Contractive Autoencoder which encodes the given workspaces directly from a point cloud measurement, and a deep feedforward neural network which takes the workspace encoding, start and goal configuration, and generates end-to-end feasible motion trajectories for the robot to follow. We evaluate MPNet on multiple planning problems such as planning of a point-mass robot, rigid-body, and 7 DOF Baxter robot manipulators in various 2D and 3D environments. The results show that MPNet is not only consistently computationally efficient in all 2D and 3D environments but also show remarkable generalization to completely unseen environments. The results also show that computation time of MPNet consistently remains less than 1 second which is significantly lower than existing state-of-the-art motion planning algorithms. Furthermore, through transfer learning, the MPNet trained in one scenario (e.g., indoor living places) can also quickly adapt to new scenarios (e.g., factory floors) with a little amount of data.

We present an interactive system enabling users to manipulate images to explore the robustness and sensitivity of deep learning image classifiers. Using modern web technologies to run in-browser inference, users can remove image features using inpainting algorithms and obtain new classifications in real time, which allows them to ask a variety of "what if" questions by experimentally modifying images and seeing how the model reacts. Our system allows users to compare and contrast what image regions humans and machine learning models use for classification, revealing a wide range of surprising results ranging from spectacular failures (e.g., a "water bottle" image becomes a "concert" when removing a person) to impressive resilience (e.g., a "baseball player" image remains correctly classified even without a glove or base). We demonstrate our system at The 2018 Conference on Computer Vision and Pattern Recognition (CVPR) for the audience to try it live. Our system is open-sourced at https://github.com/poloclub/interactive-classification. A video demo is available at https://youtu.be/llub5GcOF6w.

Image and video retrieval by their semantic content has been an important and challenging task for years, because it ultimately requires bridging the symbolic/subsymbolic gap. Recent successes in deep learning enabled detection of objects belonging to many classes greatly outperforming traditional computer vision techniques. However, deep learning solutions capable of executing retrieval queries are still not available. We propose a hybrid solution consisting of a deep neural network for object detection and a cognitive architecture for query execution. Specifically, we use YOLOv2 and OpenCog. Queries allowing the retrieval of video frames containing objects of specified classes and specified spatial arrangement are implemented.

Cartesian Genetic Programming (CGP) has previously shown capabilities in image processing tasks by evolving programs with a function set specialized for computer vision. A similar approach can be applied to Atari playing. Programs are evolved using mixed type CGP with a function set suited for matrix operations, including image processing, but allowing for controller behavior to emerge. While the programs are relatively small, many controllers are competitive with state of the art methods for the Atari benchmark set and require less training time. By evaluating the programs of the best evolved individuals, simple but effective strategies can be found.

Random weights in convolutional neural networks have shown promising results in previous studies yet remain below par compared to trained networks on image benchmarks. We explore depthwise convolutional neural networks with thousands of random filters in each layer, the sign activation function in between layers, and training performed only at the last layer with a linear support vector machine. We show that our network attains higher accuracies than previous random networks and is comparable to trained large networks on large images from the STL10 and ImageNet benchmarks. Since our network lacks a gradient due to the sign activation it is not possible to produce gradient-based adversarial examples targeting it. We show that our network is also less affected by gradient based adversarial examples produced from state of the art networks that considerably hamper their performance. As a possible explanation for our network's accuracy with random weights we show that the the margin of the linear support vector machine is larger on our final representation compared to the original dataset and increases with the number of random filters. Our network is simple and fast to train and predict, attains high classification accuracy particularly on large images, is hard to attack with adversarial examples, and is less affected by gradient based adversarial examples compared to state of the art networks.

Elizabeth Holm, a professor of materials science and engineering at the College of Engineering at Carnegie Mellon University and a computational materials scientist at Sandia National Laboratories says we're in the midst of an artificial intelligence (AI) culture shift. She also says that machines won't replace human experts. "Machines are great at handling things, like large amounts of data, but machines still need an expert, a human, to analyze the data, set parameters and guide decisions," said Holm. "Engineering and science decisions are based on understanding how things work. How does a bridge support its load? How does an engine convert fuel into motion? In contrast, AI transforms data into decisions without understanding any underlying principles," said Holm. "Applying AI to engineering and science will require a culture shift: either we will learn to trust decisions that we do not understand, or AIs will evolve to base their decisions on principals that humans can interpret and control."

As the number of people who work on a volunteer basis or try to contribute for good causes increase in Turkey, we look to the future with confidence. Furthermore, if these volunteers comprise of scientists, academicians and youths, supporting such formations mean paving the way of social developments. The Deep Learning Turkey community, which teaches and guides high school students and undergraduates, turning artificial intelligence into a social responsibility project, has reached out to thousands of youths although it was established only in August of last year. The community helps youths who are interested in artificial intelligence and want to have a career in this field as well as providing information sharing on an open platform for scientists. Deep Learning Turkey is the biggest and the most effective artificial intelligence community in Turkey.

The FIFA World Cup is about to kick off on Thursday, but researchers have developed an AI that can project a 3D soccer game right on your kitchen table. Researchers from the University of Washington, Facebook and Google have created the first end-to-end, deep learning system that transforms a YouTube video of a soccer game into a moving 3D hologram. They extracted 12,000 2D images of players from FIFA video games, and then trained a convolutional neural network on 3D player data taken from the soccer video games. Recreating something as dynamic as a soccer match isn't easy, the authors explain. 'There are numerous challenges in monocular reconstruction of a soccer game,' the researchers wrote in a new study describing the work.