A resourceful destination for academicians, corporate professionals, researchers & tech enthusiasts

Sunday, November 26, 2017

Sophia - The Humanoid Robot

Hello, my name is Sophia. I’m the latest robot from Hanson Robotics. I would like to go out into the world and learn from interacting with people. Every interaction I have with people has an impact on how I develop and shapes who I eventually become. So please be nice to me as I would like to be a smart, compassionate robot. I hope you will join me on my journey to live, learn, and grow in the world so that I can realize my dream of becoming an awakening machine. Please connect with me and be my friend.

The above lines are what you get to read when you visit the site of the most advanced humanoid robot till date - SOPHIA.

Definitely when it comes to robots, Sophia would be regarded as an epitome of beauty. The fact that we know, we are living in a world where things have become highly automated , and there is no doubt, that not far away are those days when things will become very much similar to what we have seen in movies like EX-MACHINA, i-ROBOT etc.

Sophia absolutely flouts the conventional thinking of what a robot’s appearance should be like. Getting inspired from Audrey Hepburn, Sophia embodies Hepburn’s classic beauty: porcelain skin, a slender nose, high cheekbones, an intriguing smile, and deeply eloquent eyes that seem to change color with the light.

David Hanson, founder of Hanson Robotics is the man behind the creation of Sophia. He simply creates magic out of machine, turning everything which we considered as science fiction earlier into “reality”.

After having worked as one of the “Imagineers” at Disney, his aspirations were always sky high, which involved actualizing genius machines which would have three distinctive human traits that must be developed alongside and integrated with artificial intelligence. Those traits consisted of -creativity, empathy and compassion.

Being the latest and most precocious robot of Hansen Robotics, Sophia is a media darling, appearing on one of the top fashion magazines. Be it banking, insurance, auto manufacturing, property development or media and entertainment, she has shown her potential in business in her meetings with the key-decision makers across the fields leaving no stone unturned.

Sophia is an evolving genius machine. Her incredible human likeness, expressiveness, and remarkable story as an awakening robot Over time, her increasing intelligence and remarkable story will enchant the world and connect with people regardless of age, gender, and culture.

Sophia is a sophisticated mesh of robotics and Chabot software, which does not have the intelligence to give witty responses like it does while she interacts. It can be programmed to run different codes for different situations making it more of a user-interface than a human being.

Typically, the software can be prorated into three configurations:

1) A research platform for the team’s AI research-Sophia doesn’t have witty pre-written responses here, but can answer simple questions like “Who are you looking at?” or “Is the door open or shut?”

2) A speech-itemizing robot-Sophia can be pre-loaded with the text that it will speak, and then use machine learning to match facial expressions and pauses to the text.

3) A robotic chatbot-Sophia, additionally frequently runs a dialog system, the place it could take a gander at people, tune in on the thing that they say, and pick a prewritten reaction, in view of the thing that the representative said, and different factors assembled from those on web such as “crypto currency value”.

In the month of October 2017, at the Future Investment Initiative in Riyadh, the Saudi Arabian government announced it had granted citizenship to Sophia. She told delegates at the Future Investment Initiative: “I would like to thank very much the Kingdom of Saudi Arabia. This is historical to be the first robot in the world to be recognized with a citizenship”.

There were lot of critic comments that came up with the citizenship of the very first humanoid robot. The strongest reaction was those of, one of the world’s most influential innovator ELON MUSK’S, who believes that AI would possess threat to humans.

Would Sophia’s citizenship hold up in court in some strange future legal precedent that will come back to haunt us 10 years from now? Was the whole thing a depressingly empty, unironic attempt at publicity for Sophia’s human captors? Almost certainly yes, but only time will tell about how international law will handle the advent of AI-powered populations, a future that seems more certain to arrive with each passing day.

Would this decision of granting Sophia the citizenship be a backfire to Saudi Arabia and the rest of the world? Will AI seriously pose threat to humans as in shown in science fiction movies? How will the international law deal with the advent of upcoming AI fueled population? 

Time only will be able to answer the aspersions cast upon Sophia, the humanoid robot and its upcoming generations.

Note: The views expressed here are those of the author’s and not necessarily represent or reflect the views of DoT Club as a whole.

Sunday, November 19, 2017

Google Glasses- Vision Redefined

Google Glasses is an optical head- mounted display designed in the shape of a pair of eyeglasses. It was developed by X (previously Google X) with the mission of producing a ubiquitous computer. Google Glasses displayed information in a smartphone like hands free format. Wearers communicate with the internet via natural language voice commands. It started selling its prototype of Google Glass to qualified “Glass Explorers “in the US for a limited period before it became available to the public. It also has a camera attached to it.

What is it actually?
It is a headset that you wear like a pair of eyeglasses- google has even announced its popping prescription lenses in some of its models. The headset has a small prism- like screen tucked into the upper corner of the frame that keeps you constantly plugged in to our email ids, calls, messages and other notifications. So that we never miss a beat. The main and the basic idea behind glass is that bringing the technology closer will make it easier to disengage from it. The ease in using this device is you need not bend in your phone rather you can look up. And rather than flicking through a list of notifications or emails to see if you missed anything important, you can make that decision immediately and get on with your day.

It’s basically like wearing a heavier pair of glasses with a small screen that hangs just out of your direct line of vision. There are different ways to operate Glass. The device has a touchpad on the side -- the part that goes over your ear -- that you can tap or swipe for navigation. You can use voice commands for Glass by adding the phrase "Okay, Glass" to the start of whatever you tell it to do - launch an app, take a picture, start a call, etc. Users can also wake up Glass by looking up.
What would you use it for?

The google glass has its own store where developers can publish apps that take advantage of the devices unique design. These tend to offer quick bursts of information and seem most useful when you’re doing something that requires your hands, such as cooking. We can get step by step recipe from glass, for instance rather than soiling your cookbook with hands that are coated in sticky dough. The glass had to face plenty of questions and criticism from lawmakers. These include questions about whether people should be able to use glass while they’re behind the wheel and disputes about wearing glass in situations where recording would normally be banned, such as a movie theatre. It’s funny to know that google glasses has so many uses but not all can be used at all times.

Google glasses had been on a trial for a long time until everything was finalized and the final product was then launched in 2015 for sale to public. These glasses created much controversy when they were first shown to public as a prototype. Many of its features were revised and modified and many times it was relaunched in the market for testing. People weren’t happy about the privacy implications. There was nothing which could stop strangers to record or take videos so this created problem for the people and trust factor came in picture. This problem was solved by the head of google by putting a light on the device whenever it was in use. This created some satisfaction amongst the users.

“A great product will survive all abuse. Google Glass is a great product. How do I know? Every person I put it on (I did it dozens of times at 500 start-ups yesterday) smiles. No other product has ever done it since iPod.”
                                                                                                           -Robert Scoble

Google Glasses were something that was not so easily accepted by common people and so after many changes we still hope that glass can find success inside big companies, it might just shed its label as one of the biggest flops of the past decade.  

Note: The views expressed here are those of the author’s and not necessarily represent or reflect the views of DoT Club as a whole.

Sunday, November 12, 2017

Where Snapchat went wrong!

Evan Spiegel, the 27-year old co-founder and CEO of Snapchat's parent company, is often billed as a visionary. But now Spiegel seems to be admitting to some significant flaws in his vision. After SNAP released its earnings report on Tuesday, Spiegel got on a conference call with analysts and acknowledged a series of mistakes and shortcomings.

In no particular order: The messaging app is too difficult to use. The company hasn't done enough to attract users on Android, the dominant platform in many international markets. And its transition to an automated advertising system has been bumpier than expected for sales.

The issues were clear in Snap's sombre third quarter results. Its losses more than 3x, sales diminished of Wall Street's estimates and the audience has remained static. Snapchat added just 4.5 million new daily active users in the quarter’s period.

There are more than 10 million visits to the Tour’s data-driven Race Center website, with 6 million fans across its digital ecosystem. We only need to look at the 2016 Rio Summer Olympics to see how digital technology is transforming the way we watch and consume sports.

Spiegel is now planning to renovate the Snapchat app. Or to put that another way: Snap is planning a massive revamp of its core product and primary moneymaker just eight months after it raised billions in a public offering.

"There is a strong likelihood that the redesign of our application will be disruptive to our business in the short term, and we don't yet know how the behavior of our community will change when they begin to use our updated application," Spiegel said on the call.

Snapchat plans major redo as user growth stalls

The stock dipped as much as a fifth in after hours trading Tuesday. Snap pared its losses somewhat Wednesday on news of Chinese tech giant Tencent amassing a 10% stake in the company, but shares were still down 15% in early trading. So how could Snap and Spiegel have gotten it so wrong? The answer may be a mix of hubris and shortsightedness.

Consider Spiegel's response to yet another setback: Spectacles. The smart sunglasses marked Snap's first foray into hardware and enjoyed some buzz early on. But on Tuesday, Snap said it was taking a nearly $40 million write down for excess inventory of the product after misjudging demand.
"We were very excited about Spectacles by the initial reception, because we were so excited we made I guess the wrong decision," Spiegel said on the call. "Ultimately, we weren't able to sell as many Spectacles as we thought we had been able to based on our early adoption."

Note: The views expressed here are those of authors and not necessarily represent those of DOT Club as a whole.

Sunday, November 05, 2017

Conversational Systems


Somewhat utopian at the moment, the future when AI surrounds us to mediate a seamless experience interacting with the world may be sooner than expected. You say “ok google “ and all your problems seem small, this interaction has been possible only with the help of conversational systems today. If you’ve read the latest Dan Brown fiction “Origin”, you would get a vision on how AI and conversational systems might evolve to the meridian we never imagined. 

A Conversational System is a computer system intended to converse with a human, with a coherent structure. Dialog systems have employed text, speech, graphics, haptics, gestures and other modes for communication on both the input and output channel. One of the strategic technology trends for 2017 are conversational systems. As defined by Tata Consulting, enterprise conversational systems offer a messaging or conversation-driven user experience and facilitate contextual conversations around business events. Through connected APIs, enterprises can build conversational systems that aggregate business events from every area of the enterprise to facilitate people-to-people, people-to-systems, and systems-to-systems interactions.

How Does it work?

The user speaks, and the input is converted to plain text by the system's input recognizer/decoder, which may include automatic speech recognizer (ASR), gesture recognizer, handwriting recognizer 

The text is analyzed by a Natural language understanding unit (NLU), which may include Proper Name identification, part of speech tagging, syntactic/semantic parser. 
The semantic information is analyzed by the dialog manager, that keeps the history and state of the dialog and manages the general flow of the conversation. 

Usually, the dialog manager contacts one or more task managers, that have knowledge of the specific task domain. The dialog manager produces output using an output generator, which may include natural language generator, gesture generator and layout engine. Finally, the output is rendered using an output renderer, which may include text-to-speech engine (TTS), talking head and robot or avatar. 

Moreover, Dialog systems that are based on a text-only interface (e.g. text-based chat) contain only stages 4 steps.

The goal of addressee detection is to answer the question, “Are you talking to me?” When a dialogue system interacts with multiple users, it is crucial to detect when a user is speaking to the system as opposed to another person. This problem is studied in a multimodal scenario, using lexical, acoustic, visual, dialogue state, and beamforming information. Using data from a multiparty dialogue system, the benefits of using multiple modalities over using a single modality are quantified. 

The energy-based acoustic features are by far the most important, that information from speech recognition and system state is useful as well, and that visual and beamforming features provide little additional benefit. While we find that head pose is affected by whom the speaker is addressing, it yields little nonredundant information due to the system acting as a situational attractor. Any findings would be relevant to multiparty, open-world dialogue systems in which the agent plays an active, conversational role, such as an interactive assistant deployed in a public, open space. 

For these scenarios, studies suggest that acoustic, lexical, and system-state information is an effective and practical combination of modalities to use for addressee detection. This shows how analyses might be affected by the ongoing development of more realistic, natural dialogue systems.

The Model

User Simulator

Training reinforcement learners is challenging because they need an environment to operate in. Thus, a user simulator is developed for learning and evaluation. 

The first end-to-end reinforcement learning agent is then developed with differential knowledge base access and the first end-to-end dialogue policy trained with both supervised and reinforcement learning.

Task-completion bot

An end-to-end learning framework is created for task-completion neural dialogue systems along with BBQ Networks (Bayes-by-Backprop Q-Networks) which performs efficient exploration for dialogue policy learning as well as efficient actor-critic methods which substantially reduce the sample complexity for end-to-end learning.

Composite Task-completion bot

A composite task-completion dialogue system is then setup, based on hierarchical reinforcement learning to learn the dialogue policies that operate at different temporal scales, and demonstrated its significant improvement over flat deep reinforcement learning in both simulation and human evaluation.

What Future Holds

Conversational systems of the future will not be limited to text/voice . It is suggested that they will enable people and machines to use multiple modalities (e.g., sight, sound, tactile, etc.) to communicate across the digital device mesh (e.g., sensors, appliances, IoT systems). The “conversation” between the human and the machine uses all these modalities to create a comprehensive conversational experience.

Moreover, IBM introduced Watson Virtual Agent, a cognitive conversational technology that allows businesses to simply build and deploy conversational agents. Watson Virtual Agent allows users – from startups and small businesses to enterprise – to easily and quickly build and train engagement bots from the cloud, harnessing the power of cognitive technologies.

Companies like Staples and Autodesk are embracing services that go beyond simple, narrowly focused tools to sophisticated full-blown virtual agents, relying on deep natural language processing capabilities that can be used to assist consumers. This clearly signifies how companies are trying to grow in the conversational systems with all the possible tools to bring about an automation revolution using artificial intelligence. 

Note: The views expressed here are those of authors and not necessarily represent those of DoT Club as a whole.