My newest Ask-A-Data-Scientist post was inspired by a computer science student who wrote in asking for advice on how to pursue a career in policy making related to the societal impacts of AI. I realized that there are many great resources out there, and I wanted to compile a list of links all in one place. You can find my previous Ask-A-Data-Scientist advice columns here. Everyone in tech should be concerned about the ethical implications of our work and actively engaging with such questions. The humanities and social sciences are incredibly relevant and important in addressing ethics questions.

Hyperparameter optimization can be formulated as a bilevel optimization problem, where the optimal parameters on the training set depend on the hyperparameters. We aim to adapt regularization hyperparameters for neural networks by fitting compact approximations to the best-response function, which maps hyperparameters to optimal weights and biases. We show how to construct scalable best-response approximations for neural networks by modeling the best-response as a single network whose hidden units are gated conditionally on the regularizer. We justify this approximation by showing the exact best-response for a shallow linear network with L2-regularized Jacobian can be represented by a similar gating mechanism. We fit this model using a gradient-based hyperparameter optimization algorithm which alternates between approximating the best-response around the current hyperparameters and optimizing the hyperparameters using the approximate best-response function. Unlike other gradient-based approaches, we do not require differentiating the training loss with respect to the hyperparameters, allowing us to tune discrete hyperparameters, data augmentation hyperparameters, and dropout probabilities. Because the hyperparameters are adapted online, our approach discovers hyperparameter schedules that can outperform fixed hyperparameter values. Empirically, our approach outperforms competing hyperparameter optimization methods on large-scale deep learning problems. We call our networks, which update their own hyperparameters online during training, Self-Tuning Networks (STNs).

Heterogeneous knowledge naturally arises among different agents in cooperative multiagent reinforcement learning. As such, learning can be greatly improved if agents can effectively pass their knowledge on to other agents. Existing work has demonstrated that peer-to-peer knowledge transfer, a process referred to as action advising, improves team-wide learning. In contrast to previous frameworks that advise at the level of primitive actions, we aim to learn high-level teaching policies that decide when and what high-level action (e.g., sub-goal) to advise a teammate. We introduce a new learning to teach framework, called hierarchical multiagent teaching (HMAT). The proposed framework solves difficulties faced by prior work on multiagent teaching when operating in domains with long horizons, delayed rewards, and continuous states/actions by leveraging temporal abstraction and deep function approximation. Our empirical evaluations show that HMAT accelerates team-wide learning progress in difficult environments that are more complex than those explored in previous work. HMAT also learns teaching policies that can be transferred to different teammates/tasks and can even teach teammates with heterogeneous action spaces.

Federated Learning allows multiple parties to jointly train a deep learning model on their combined data, without any of the participants having to reveal their local data to a centralized server. This form of privacy-preserving collaborative learning however comes at the cost of a significant communication overhead during training. To address this problem, several compression methods have been proposed in the distributed training literature that can reduce the amount of required communication by up to three orders of magnitude. These existing methods however are only of limited utility in the Federated Learning setting, as they either only compress the upstream communication from the clients to the server (leaving the downstream communication uncompressed) or only perform well under idealized conditions such as iid distribution of the client data, which typically can not be found in Federated Learning. In this work, we propose Sparse Ternary Compression (STC), a new compression framework that is specifically designed to meet the requirements of the Federated Learning environment. Our experiments on four different learning tasks demonstrate that STC distinctively outperforms Federated Averaging in common Federated Learning scenarios where clients either a) hold non-iid data, b) use small batch sizes during training, or where c) the number of clients is large and the participation rate in every communication round is low. We furthermore show that even if the clients hold iid data and use medium sized batches for training, STC still behaves pareto-superior to Federated Averaging in the sense that it achieves fixed target accuracies on our benchmarks within both fewer training iterations and a smaller communication budget.

Natural Language Processing (NLP) is a hot topic into Machine Learning field. This course is an advanced course of NLP using Deep Learning approach. Before starting this course please read the guidelines of the lesson 2 to have the best experience in this course. This course starts with the configuration and the installation of all resources needed including the installation of Tensor Flow CPU/GPU, Cuda and Keras. You will be able to use your GPU card if you have one, to accelate so fast the processes.

We capture millions of data points on a daily basis, a human could not process that many data points,

Google and online learning hub Udacity have launched a free course designed to make it simpler for software developers to grasp the fundamentals of machine learning. The "Intro to TensorFlow for Deep Learning" course is designed to be more accessible to developers than previous machine-learning courses offered by Udacity. "Our goal is to get you building state-of-the-art AI applications as fast as possible, without requiring a background in math," says Mat Leonard, head of the School of AI at Udacity. "If you can code, you can build AI with TensorFlow. You'll get hands-on experience using TensorFlow to implement state-of-the-art image classifiers and other deep learning models. You'll also learn how to deploy your models to various environments including browsers, phones, and the cloud."

I will be the first to admit that certain questions have no right or wrong answers. An example is, will artificial intelligence or even technology in general make us more or less intelligent? Looking at it from a broader perspective, one cannot deny the obvious that technology has impacted the society for good, but at the same time, it has some negative sides that we have to deal with. Crop improvement, genetics, three dimensional technology, blockchain and many more are some of the positives that we have gained from the advancement of technology but some activities from the processes that gave us these technological breakthroughs, such as pollution which creates environmental hazards, has given technology some negatives in the view of many of its sceptics. Today, AI, (one of the daring outcomes of continuous technological advancement) is arguably one of the most discussed trends in the world of technology, mainly on how it is helping to improve different aspects of society evolvement.

Some of the nation's top schools, including the Massachusetts Institute of Technology and Georgetown University, are offering educational programs for nontechnical senior managers looking to learn more about artificial intelligence and its business applications. These programs are designed to be crash courses in AI, covering machine learning, robotics and other topics. They can vary in duration and delivery--some are multiweek online offerings, others are multiday boot camps--and typically land in the $3,000-a-student range.

Google has kept its education initiatives, ranging from Chromebooks for schools to teaching software, within the confines of the classroom. "Bolo" (Hindi for "speak") is essentially a reading assistant for elementary school children that harnesses Google's speech recognition and text-to-speech smarts. Kids can read through a mix of Hindi and English stories with the help of an in-app learning assistant, who appears as a young girl named "Diya." She offers tips and corrections and can even translate Hindi words into English. The app supports multiple users and works offline in order to avoid connectivity issues in rural areas -- while Google works on getting those regions online through WiFi hotspots in remote train stations.