Since the early days of machine learning, artificial intelligence scenarios have faced with two big challenges in order to experience mainstream adoption. First, we have the data efficiency problem that requires machine or deep learning models to be trained using large and accurate datasets which, as we know, are really expensive to build and maintain. Secondly, we have the generalization problem which AI agents face in order to build new knowledge that is different from the training data. Humans, by contrast, are incredibly efficient learning with minimum supervision and rapidly generalizing knowledge from a few data examples. Generative models are one of the deep learning disciplines that focuses on addressing the two challenges mentioned above.

Information about the built environment at scale provides actionable insight that can transform a broad spectrum of industries, by enabling companies to deliver better, more targeted services to customers, telecommunications operators to perform accurate infrastructure planning, and government agencies to conduct regional planning and management. This is the vision behind Geoscape, a location-intelligence product conceived by PSMA Australia and developed in partnership with DigitalGlobe, a Maxar company, that delivers detailed information about buildings and their attributes, trees and land cover for every address in Australia. The data for each building in Geoscape includes the presence or absence of solar panels. This information is important for the insurance industry because solar panels may pose a fire hazard and thus affect insurance premiums. For firefighters, knowing if a structure has solar panels can also help them stay safe.

Deep learning has revolutionised many fields, but it is still challenging to transfer its success to small mobile robots with minimal hardware. Specifically, some work has been done to this effect in the RoboCup humanoid football domain, but results that are performant and efficient and still generally applicable outside of this domain are lacking. We propose an approach conceptually different from those taken previously. It is based on semantic segmentation and does achieve these desired properties. In detail, it is being able to process full VGA images in real-time on a low-power mobile processor. It can further handle multiple image dimensions without retraining, it does not require specific domain knowledge for achieving a high frame rate and it is applicable on a minimal mobile hardware.

You've probably heard the phrase'deep learning' bandied about in conversation, or maybe you've read about it in a post like this one. It seems like almost every tech conversation happening today somehow touches on the topic of AI, machine learning, or deep learning.

Researchers have developed a fully integrated computer-aided diagnosis (CAD) system that detects, segments and classifies masses from mammograms using deep learning and a deep convolutional neural network (CNN), according to a new study published by the International Journal of Medical Informatics. The authors used a proven regional deep learning model, You-Only-Look-Once (YOLO), for detecting potential masses in the breast tissues of patient data taken from a publicly available dataset. They provided a number of reasons in their analysis. "First, YOLO has a robust ability to detect the masses directly from entire mammograms," wrote author Mugahed A. Al-antari, PhD, of Kyung Hee University in Yongin, South Korea, and colleagues. "Second, detected bounding boxes via YOLO accurately align the masses, thereby, a low rate of false positives is achieved compared with other studies. Third, it can even detect challenging cases where the masses exist either over pectoral muscles or inside dense regions. Fourth, the running time of the testing and required memory are extremely low compared to other more complex deep learning models."

In this tutorial, we will demonstrate how to implement the state of the art End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF paper for Named Entity Recognition using Pytorch. The main aim of the tutorial is to make the audience comfortable with Pytorch using this tutorial and give a step-by-step walkthrough of the Bi-LSTM-CNN-CRF architecture for Named-Entity-Recognition.

AI, machine learning, and deep learning - these terms overlap and are easily confused, so let's start with some short definitions. AI means getting a computer to mimic human behavior in some way. Machine learning is a subset of AI, and it consists of the techniques that enable computers to figure things out from the data and deliver AI applications. Deep learning, meanwhile, is a subset of machine learning that enables computers to solve more complex problems. Those descriptions are correct, but they are a little concise.

Passage AI, a leader in AI-powered conversational interfaces, announced it has extended its natural language understanding/processing (NLU/P) capabilities to support over 150 languages natively, including simplified and traditional Chinese, French, German, Portuguese, Japanese and Spanish. Companies can now offer near instant responses to customers around the globe in their native tongue. "Passage AI-powered chatbots now enable 24/7 customer service responses for routine questions in all major languages, delivering a superior customer experience for consumers regardless of the language they speak," said Ravi N. Raj, CEO and co-founder of Passage AI. "The world is flat, so it is not surprising that our customers, some of the largest brands in the world, want support for multiple languages. Our customers have reported 95 percent or higher accuracy in field tests, which analysts have said is the leading benchmark for AI-powered chatbots in any industry, but particularly important for customer-facing interfaces." Market research firm Vanson Bourne, in a survey sponsored by Teradata, found that 80 percent of organizations are currently investing in AI, and on average each company has spent $6-8 million so far.

The non-stationarity characteristic of the solar power renders traditional point forecasting methods to be less useful due to large prediction errors. This results in increased uncertainties in the grid operation, thereby negatively affecting the reliability and increased cost of operation. This research paper proposes a unified architecture for multi-time-horizon predictions for short and long-term solar forecasting using Recurrent Neural Networks (RNN). The paper describes an end-to-end pipeline to implement the architecture along with the methods to test and validate the performance of the prediction model. The results demonstrate that the proposed method based on the unified architecture is effective for multi-horizon solar forecasting and achieves a lower root-mean-squared prediction error compared to the previous best-performing methods which use one model for each time-horizon. The proposed method enables multi-horizon forecasts with real-time inputs, which have a high potential for practical applications in the evolving smart grid.

Humans are ingenious: We have unique abilities to predict the consequences of observed actions, remember the most relevant experiences from the past, and transfer knowledge from previous observations in order to adapt to novel situations. The episodic memory which encodes contextual, spatial and temporal experiences during development plays a vital role to introduce such cognitive abilities in humans. A core challenge in cognitive robotics is compact and generalizable mechanism which allow for encoding, storing and retrieving spatiotemporal patterns of visual observations. Such mechanisms would enable robots to build a memory system, allowing them to efficiently store gained knowledge from past experiences and both recalling and applying such knowledge in new situations. Inspired by infants that learn by observing and memorizing what adults do in the same visual setting, we investigate in this paper how to extend cognitive abilities of robots to autonomously infer the most probable behavior and ultimately adapt it to the current scene. Considering the situation of the humanoid robot ARMAR-IIIa standing in front of a table with a juice carton (see Figure 1) one can ask what the most suitable action is and how it would best be performed. To achieve this goal, we introduce a novel deep neural network architecture for encoding, storing, and recalling past action experiences in an episodic memory-like manner.