This post contains a list of the AI-related seminars that are scheduled to take place between 15 March and 30 April 2021. All events detailed here are free and open for anyone to attend virtually. Scalable robot learning in rich environments Speaker: Raia Hadsell (DeepMind) Organised by: Carnegie Mellon University Zoom link is here. Adaptive estimator selection for off-policy evaluation Speaker: Yi Su (Cornell) Organised by: RL theory The seminar will be livestreamed here. Title to be confirmed Speaker: Munther Dahleh (MIT) Organised by: Control Meets Learning Watch here.

Deep learning systems are revolutionizing technology around us, from voice recognition that pairs you with your phone to autonomous vehicles that are increasingly able to see and recognize obstacles ahead. But much of this success involves trial and error when it comes to the deep learning networks themselves. A group of MIT researchers recently reviewed their contributions to a better theoretical understanding of deep learning networks, providing direction for the field moving forward. "Deep learning was in some ways an accidental discovery," explains Tommy Poggio, investigator at the McGovern Institute for Brain Research, director of the Center for Brains, Minds, and Machines (CBMM), and the Eugene McDermott Professor in Brain and Cognitive Sciences. "We still do not understand why it works. A theoretical framework is taking form, and I believe that we are now close to a satisfactory theory. It is time to stand back and review recent insights."

The choice between traditional machine vision and deep learning depends upon the type of application being solved, the amount of data being …

Computer vision has been evolving very rapidly in the last decade through deep learning [3] and is revolutionizing businesses around the globe. Benefits of computer vision can be realized across industries and along the entire value chain of a company. Let's consider a few examples of industry segments impacted by computer vision. Visual inspection can help in the manufacturing industry in ways such as quality inspection, product development, security, surveillance, and worker safety. In Table 1, we compare traditional visual inspection with the computer vision approach for visual inspection.

So far we have discussed linear regression and gradient descent in previous articles. We got a simple overview of the concepts and a practical tutorial to understand how they work. In this article, we will see the mathematics behind gradient descent and how can an "optimizer" get the global minima point. If the term "optimizer" is new for you, it is simply the function that works to determine the global minima point which refers to the coefficients of best-fit line in linear regression algorithm. By the way, similar concepts are used in deep learning algorithms.

In general, Graph Neural Networks (GNN) refer to the general concept of applying neural networks (NNs) on graphs. In a previous article, we cover GCN which is one of the popular approaches in GNN. But in some literature, GNN may refer to a more specific approach that the hidden state of a node depends on its last states and its neighbors. We can view this as message passings with its neighboring nodes. In this article, we will look into some of its specific implementations.

Graph Neural Networks is a neural network architecture that has recently become more common in research publications and real-world applications. And since neural graph networks require modified convolution and pooling operators, many Python packages like PyTorch Geometric, StellarGraph, and DGL have emerged for working with graphs. In Keras Graph Convolutional Neural Network(kgcnn) a straightforward and flexible integration of graph operations into the TensorFlow-Keras framework is achieved using RaggedTensors. It contains a set of TensorFlow-Keras layer classes that can be used to build graph convolution models. The package also includes standard bench-mark graph datasets such as Cora,45 MUTAG46, and QM9.

With a humongous 2.5 quintillion bytes of data produced every day, data scientists are busier than at any other time. The more data we have, the more we can do with it. Furthermore, data science gives us strategies to effectively utilize this data. It just bodes well that software engineering has developed to incorporate data engineering skill, a subdiscipline that focuses on the transportation, change, and storage of data. Data engineering is a subset of data science, a comprehensive term that incorporates numerous fields of information related to working with data.

The recent remarkable progress of deep reinforcement learning (DRL) stands on regularization of policy for stable and efficient learning. A popular method, named proximal policy optimization (PPO), has been introduced for this purpose. PPO clips density ratio of the latest and baseline policies with a threshold, while its minimization target is unclear. As another problem of PPO, the symmetric threshold is given numerically while the density ratio itself is in asymmetric domain, thereby causing unbalanced regularization of the policy. This paper therefore proposes a new variant of PPO by considering a regularization problem of relative Pearson (RPE) divergence, so-called PPO-RPE. This regularization yields the clear minimization target, which constrains the latest policy to the baseline one. Through its analysis, the intuitive threshold-based design consistent with the asymmetry of the threshold and the domain of density ratio can be derived. Through four benchmark tasks, PPO-RPE performed as well as or better than the conventional methods in terms of the task performance by the learned policy.

As deep neural networks (DNNs) become deeper, the training time increases. In this perspective, multi-GPU parallel computing has become a key tool in accelerating the training of DNNs. In this paper, we introduce a novel methodology to construct a parallel neural network that can utilize multiple GPUs simultaneously from a given DNN. We observe that layers of DNN can be interpreted as the time steps of a time-dependent problem and can be parallelized by emulating a parallel-in-time algorithm called parareal. The parareal algorithm consists of fine structures which can be implemented in parallel and a coarse structure which gives suitable approximations to the fine structures. By emulating it, the layers of DNN are torn to form a parallel structure, which is connected using a suitable coarse network. We report accelerated and accuracy-preserved results of the proposed methodology applied to VGG-16 and ResNet-1001 on several datasets.