Accelerating Deep Learning Training with BigDL and Drizzle on Apache Spark - RISE Lab


This work was done in collaboration with Ding Ding and Sergey Ermolin from Intel. In recent years, the scale of datasets and models used in deep learning has increased dramatically. Although larger datasets and models can improve the accuracy in many AI applications, they often take much longer to train on a single machine. However, it is not very common to distribute the training to large clusters using current popular deep learning frameworks, compared to what's been long around in the Big Data area, as it's often harder to gain access to a large GPU cluster and lack of convenient facilities in popular DL frameworks for distributed training. By leveraging the cluster distribution capabilities in Apache Spark, BigDL successfully performs very large-scale distributed training and inference.

Artificial intelligence has come of age: R Chandrashekhar


HYDERABAD: As Saudi Arabia's decision to grant citizenship to a robot makes international headlines, a top official of the Indian IT industry body says it signifies that artificial intelligence (AI) has come of age. The AI is fast reaching a stage where many of the tasks performed by humans can be done by robots, Nasscom president R Chandrashekhar told. Riyadh's recent move to give citizenship to an AI humanoid robot is more of a "symbolic gesture" meant to draw the attention to the use of technology and willingness to give technology a free play, he said. "Obviously it (granting citizenship to a human robot, named Sophia) was not required; it's given more as a symbolic gesture," Chandrashekhar said, adding that, however, "there are lot of ethical questions which come up". He said: "But essentially, the grant of citizenship (to a bot) is only a sort of symbolic thing.

Let your car tell you what it needs

MIT News

Imagine hopping into a ride-share car, glancing at your smartphone, and telling the driver that the car's left front tire needs air, its air filter should be replaced next week, and its engine needs two new spark plugs. Within the next year or two, people may be able to get that kind of diagnostic information in just a few minutes, in their own cars or any car they happen to be in. They wouldn't need to know anything about the car's history or to connect to it in any way; the information would be derived from analyzing the car's sounds and vibrations, as measured by the phone's microphone and accelerometers. The MIT research behind this idea has been reported in a series of papers, most recently in the November issue of the journal Engineering Applications of Artificial Intelligence. The new paper's co-authors include research scientist Joshua Siegel PhD '16; Sanjay Sarma, the Fred Fort Flowers and Daniel Fort Flowers Professor of Mechanical Engineering and vice president of open learning at MIT; and two others.

The Big Data Boom Automobile Magazine


With the help of Microsoft, last year Toyota created a new data analytics division called Toyota Connected to bring Internet-connected services into the car. Earlier this year, Renault-Nissan inked a deal to leverage Microsoft's Connected Vehicle Platform and its Azure cloud architecture to collect vehicle sensor and usage data in order to develop "connected driving experiences." Ford recently invested $182 million in Pivotal, a cloud-based software company, in part to create analytics tools and a cloud platform to support the automaker's Smart Mobility initiative. Cadillac introduced the first production vehicle-to-vehicle communication system on its 2017 models, and last year, Audi launched a Traffic Light Information vehicle-to-infrastructure system that lets its cars know how long a light will stay red or green to help improve traffic flow.

Evaluating Data Science Projects: A Case Study Critique


By convention, the rare class is usually positive, so this means the True Positive (TP) rate is 0.78, and the False Negative rate (1 – True Positive rate) is 0.22. The Non-Large Loss recognition rate is 0.79, so the True Negative rate is 0.79 and the False Positive (FP) rate is 0.21. They don't report a False Positive rate (or True Negative rate, from which we could have calculated it). This result means that, using their Neural network, they must process 28 uninteresting Non-Large Loss customers (false alarms) for each Large-Loss customer they want.

How Automotive AI Is Going to Disrupt (Almost) Every Industry - DZone AI


SAE International has created the now-standard definitions for the six distinct levels of autonomy, from Level 1 representing only minor driver assistance (like today's cruise control) to Level 6 being the utopian dream of full automation: naps and movie-watching permitted. Many of the features of AI-assisted driving center around increased safety, like automatic braking, collision avoidance systems, pedestrian and cyclists alerts, cross-traffic alerts, and intelligent cruise control. A connected vehicle could also share performance data directly with the manufacturer (called "cognitive predictive maintenance"), allowing for diagnosis and even correction of performance issues without a stop at the dealer. Although it may not at first appear directly tied to automotive AI, the health and medical industry stands to experience some significant disruptions as well.

Robot is the Boss: 4 Ways You'll Soon Be Working With Robots


Artur Kiulian is the author of, Robot Is The Boss: How To Do Business with Artificial Intelligence, a book that teaches companies how to incorporate machine learning into their businesses. In the case of frequent violation of Uber algorithmic policies, the driver's account may be deactivated. Whenever a company transfers responsibility for decision-making to ML algorithms, machines automatically turn into bosses. Educating ourselves by reading books like Kiulian's Robot Is The Boss: How To Do Business with Artificial Intelligence will begin to prepare us for the reality that is already here and expanding in new ways daily.

Why Your Next Boss Will Be A Robot – Hacker Noon


In this model, human employees augment and assist AI software leveraging its natural language processing (NLG), analytics, image recognition or other ML functionality to run business processes and make important decisions. Companies can also leverage the power of complex image recognition software to automatically assess the performance of employees in contexts where it can not be properly measured by human supervisors. AI software is in charge of important business decisions, planning and performance assessment in many on-demand mobility and delivery services that make up the so called gig economy. Deliveroo's algorithmic system carefully monitors a courier's performance calculating his/her average "time to accept orders", "travel time", and "unassigned orders".

Finding Harmony Between Human and Artificial Intelligence


HUMAN INTELLIGENCE HAS ALWAYS PLAYED A CRITICAL ROLE IN MACHINE LEARNING Specifically, in supervised learning, human intelligence is generally applied to assign labels, or richer annotations, to examples used for training AI models which are then deployed in fully automated systems. In the case of an artificial agent fielding calls, text messages or other inputs from a user, human intelligence can be engaged in real time to provide live supervision of the behavior of the automated solution at various different levels (Human-assisted AI). In addition, human decisions to adopt, reject, or edit suggested responses provide critical feedback for improvement of the AI models making the suggestions. Perhaps the best solutions for customer care will combine both humans assisting AI and AI assisting humans: Customers will first engage with automated virtual assistants that respond to their calls, texts, messages and other inputs, and human assistance will play a role in optimizing performance.

How big data is transforming the automotive industry


By 2020, the connected car market report states that connected car services will account for approximately $40 billion annually. These services include infotainment, navigation, fleet management, remote diagnostics, automatic collision notification, enhanced safety, usage based insurance, traffic management and, lastly, autonomous driving. The root of these applications is big data, as increasing amounts of data are collected from remote sensors; this information is being interpreted and leveraged to transform the automotive industry into one of automation and self-sufficiency. By using the information gleaned from smart sensors, the industry can benefit from compiling custom insurance plans, monitoring driver behaviour, performance and safety.