Monthly Archives: July 2017

I was amused reading the recent back and forth between Facebook's (NASDAQ:FB) Mark Zuckerberg and Tesla's (NASDAQ:TSLA) Elon Musk. It is funny when two heavy hitters, both of whom arguably should have a good understanding about a given piece of technology, fall on such opposite ends of the spectrum about its capabilities. Essentially it all started when someone asked Zuckerberg what he thought of Musk and his warnings about AI being an existential threat to humanity. He responded saying it was very irresponsible, to which Elon Musk replied on Twitter that Mark has limited knowledge about the subject. Is one of them wrong and the other right? Yes, I believe so, but more about that later. This exchange also got me thinking, is this belief the source of the lot of claims Musk has made about the capabilities and development of Tesla's Autopilot system? In this article, I will look into this a little more and show you some of the challenges faced in computer vision that illustrate the difficulties faced by a primarily vision-based fully autonomous driving system like Tesla's Autopilot 2.0.

It was just over six months back that Elon Musk promised customers would start seeing FSD features in Tesla's AP2 system over the next six months at the latest. However, customers haven't even seen their "Enhanced" Autopilot match the performance of the previous version of Autopilot let alone see any unique FSD features.

Now for anyone familiar with the technology, the fact that Tesla is facing a lot of challenges comes as no surprise. Unless Tesla makes significant strides beyond the current, state-of-the-art computer vision, I do not believe there is any way it can truly have a competent fully autonomous driving system under its current Autopilot 2.0 configuration. However, this is exactly what Tesla has been "promising" going as far as to hint that it would have details on a fully autonomous ride-sharing platform before the end of this year. By the way, that text hasn't changed since last year, so I guess we should now expect these details by end of 2018?

Source: Tesla

For a long time now, I had said Elon Musk and Tesla have been making knowingly misleading statements about the potential future capabilities of their system. However, as I hear his views more and more, I am starting to question if he really understands the technology and how it works. Before I go into this, let me take a step back and talk a bit about what has caused the recent revolution in the field of AI.

People have been using computers to automate a lot of work that humans have traditionally had to do manually. Indeed computers now encompass all aspects of human life. However, there were some problems that computer programmers found hard to write solutions for. One of the best examples of this is pattern recognition for example in the form of computer vision.

I can easily spot my wife in a large crowd of people, but if someone asks me how I do it, I will find it hard to describe what features I rely on to identify her. This is also one of the reasons witnesses find it hard to describe a potential suspect without the help of a sketch artist. It is generally hard to codify how the human brain interprets patterns from such visual data. This is where the idea of deep learning with large Convolutional Neural Networks comes in handy. The basic idea is that rather than trying to identify what features the system should look for, you instead just feed the system lots of data and have it learn the features that will help it identify similar objects.

The only problem here was that it took an excruciatingly long time to train a complex "Deep" network using large training datasets. With the advent of the use of GPU processing in training these systems, we are now able to use large amounts of data to train them in a reasonable amount of time. A job that would take weeks to months of running on a large CPU cluster earlier is now possible in a few hours to a few days, and this vastly expands the practical applications of these kinds of systems in pattern recognition. Over at FundamentalSpeculation, our price action based Momentum model is a result of harnessing this power of GPU processing in training.

However, at the end of the day, it is important to understand that pattern recognition is all that these systems are doing. It is not that they suddenly have a deeper understanding of what the objects in the picture represent. This is well illustrated for example by the work of Papernot, et al. in a recent paper where they used an adversarial system to trick a classifier that identifies road signs from MetaMind. This is an instance where there is an adversarial system being used specifically to target the classifier. But also consider edge cases where there is graffiti or a sticker or something similar on the sign and you realize there can be more benign instances where this can become a problem.

Source: Papernot, et al. The image on the left is an original image of a stop sign. The image on the right is a perturbed image that causes the classifier to mis-classify it as a yield sign even though it looks the same to the human eye.

To perform a real-world and properly-blinded evaluation, we attack a DNN hosted by MetaMind, an online deep learning API. We find that their DNN misclassifies 84.24% of the adversarial examples crafted with our substitute. -Papernot, et al.

The same concept also applies to achievements in the field of deep reinforcement learning. A lot of people including me celebrated when DeepMind's AlphaGo beat the world's best Go player earlier this year. However, even in this example, if the size of the board was any different than what the system was trained on, it would have failed miserably.

Humans have a tendency to see a machine perform a particular task and think of it as having a similar level of competency to a human capable of performing the same task. This generalization cannot be applied to the field of Machine Learning/AI. To be clear, I am not saying automation will not cause massive disruption to the economy. It has the potential to displace a lot of jobs. However, the idea that applications of "Narrow" AI that are becoming prevalent today are an existential risk and need to be regulated is silly. It is also the reason why "billions of autopilot miles" is a silly metric to gauge any potential advantage Tesla may have over its competition.

Driving a car is a very complex task. The reason the average human is reasonably good at it is because humans have "intelligence" which allows them to have a high level understanding of their environment and handle most of the edge cases fairly easily. This is fundamentally not true for AI systems we have today. When you think of these systems in terms of "automation" rather than "intelligence", you start to realize some of the hard challenges that need to be overcome. To top it all, Tesla's attempt to try to achieve this using a primarily vision-based system if anything puts it at a significant disadvantage to the rest of the competition.

So what does this mean for Tesla's Autopilot system? Firstly, a big part of my critique about its design is that it has minimal redundancy in its system. It is primarily a vision based system. However, this idea is sometimes challenged by stating that it has multiple cameras that offer redundancy. I don't mean to pick on my fellow contributor ValueAnalyst, but this was the most recent exchange in writing I remember having about this topic in a recent article I published about Audi's (OTCPK:AUDVF) new Level 3 system.

He is not alone in making this assumption. There was also a recent paper claiming why adversarial attacks like the one I talked about earlier do not apply to autonomous vehicles because there is redundancy in the various angles and scales at which the same object is observed and this defeats instances of mis-classification. What I find really funny is the rebuttal to this argument came from Musk's very own OpenAI.

We've created images that reliably fool neural network classifiers when viewed from varied scales and perspectives. This challenges a claim from last week that self-driving cars would be hard to trick maliciously since they capture images from multiple scales, angles, perspectives, and the like. - OpenAI Blog

Tesla's challenge though is even bigger. What we have been talking about so far is classification of well-defined categories of objects. Tesla's system however not only has to identify and classify these well-defined objects but also reliably identify any other potential object that is an obstacle in its drive path without false positives. This makes it a significantly harder problem. Again, I'm not saying the systems won't improve. However, the current state-of-the-art systems are not reliable enough to do this, and there is no reason to believe Tesla has surpassed the state of the art in this field. This makes Tesla's claim of achieving full autonomy in this short time frame all the more ridiculous.

None of this would matter that much if everyone understood that this as a field of research with future potential and not factor in its impact into current valuations. However, this is not the case. Again, I don't want to pick on any one person, but I'm drawn to Morgan Stanley's Adam Jonas. He is currently neutral on the stock after it reached his price target, but let's take a look at what got him to that price target. I remember watching one of his interviews on CNBC a few months back and having a hard time stopping myself from throwing stuff at the television to stop the insanity. What was really amazing about his performance was his confidence in talking about a topic he clearly has no understanding about. I highly recommend you watch his full interview.

We continue to believe over 100% of the upside from the current price to our $305 target can be accounted for by the value of Tesla Mobility, an on-demand and highly automated transportation service we anticipate to be launched at low volume in 2018. - Morgan Stanley

In a note he sent clients earlier that month, he tried to break down the prospects of the company by segments. He had assigned zero value to Tesla Energy because of its negative margins and he believed any prospects for this segment would be a rounding error in the grand scheme of things. He was significantly below the Street and management on the sales volume for the Model 3 while being higher than most on the average selling price of the Model 3 ($60,000). He spoke about some of his assumptions in an interview with Bloomberg at the time. He also is not all that very bullish on the sale of electric vehicles to individual customers. He gets to his valuation based on the assumption that Tesla will have a deployable autonomous driving system that can be used for ride-sharing within a Tesla Network.

"Well, we think the electric cars for private use really are ... for human driving pleasure for wealthier individuals. That's why it's so important that in the shared model where you're not driving 10,000 miles a year, but 50 or 100 in a fleet operation, then the economics of electrification you can get that pay back period under three years. That's the game changer - shared." - Adam Jonas, Morgan Stanley

So now what happens if this possibility evaporates? How much real demand would there be for a $35,000 electric vehicle if the possibility of generating revenue via participating in an autonomous ride-sharing network goes away? Forget even that, how many people are really interested in a midsize luxury sedan if the convenience of autonomous driving is taken away? Consider the volume of all Small/Midsize Luxury car sales in America last year shown below:

Source

The entire Small/Midsize Luxury segment sold a little over 800,000 vehicles across all manufactures and models in the US in 2016. Without the prospects of autonomous driving, this will be the target market for the Model 3. Further, as other manufacturers start to bring some of their more advanced driver assist and limited self-driving features to lower-range models, Tesla, relying primarily on its vision-based system, will have a hard time keeping up with the functionality offered by its competitors.

For a long time now, I have believed Elon Musk has been misleading TSLA shareholders and customers about the potential and development curve of Tesla's Autopilot system. While I still believe that, I am now starting to question whether he genuinely understands the capabilities and limitations of narrow AI systems. At the end of the day, neither option is good for Tesla shareholders. If and when the Model 3 starts to ship in large volumes, those future customers will have a lot of expectations for the autonomous driving technology promised to them and they may not be as forgiving as some of the early adopters have been so far.

Disclosure: I am/we are short TSLA.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: The Content provided in this article should be used for informational and educational purposes only and is not intended to provide tax, legal, insurance, investment, or financial advice, and the content is not intended to be a substitute for professional advice. Always seek the advice of a relevant professional with any questions about any financial, legal or other decision you are seeking to make. Any views expressed by Laxman Vembar are his own and do not necessarily reflect the view, opinions and positions of FundamentalSpeculation.IO. Finally, you should not rely solely on the information provided by the models on FundamentalSpeculation.IO in making investment decisions, but you should consider this information in the context of all information available to you.

See original here:

Automation != Intelligence - Seeking Alpha

I love watching my son mastera LEGO set. Logic is the key ingredient in putting a set together, especially the super complexmodels that require many hours to assemble.

The LEGO company is smart. For the complex sets, the company breaks the whole into units, which are packaged to be built separately. The assembled sections are then added together to create the overall object.

The unit approach makes it much less overwhelming, while also making it easier to find the right pieces. It also helps kids realize success as each section is completed.

An added bonus is that the sections become a pathway to producing something different and creative. Many times my son has reused those units by putting togethermodules from a variety of sets to create something different and original,thereby expanding his enjoyment of the toy.

Marketing automation has a lot in common with building LEGOs. Complexmarketing automation campaigns designed and built unit by unit areeasier and more effective than anentire campaign implemented as a single unit.

From a psychological standpoint, smaller modules are easier to focus on than one large overwhelming campaign. By breaking down a campaign into small units, it becomes much easier to tackle and the focused-brain can hone in on exactly what is needed, as Eric Ravenscraft wrote in Lifehacker.

Designing modules within the marketing automation platform can lead to reuse. Different subsets can be used in different campaigns.

The units (for example a campaign flow or contact filter) can either be copied and the new version slightly updated or they can simply be used as they are. This saves time when building new campaigns.

Testing is also simplerformodularly-designed campaigns. Spotting and fixing issues at a unit level is easier than at the overall campaign level when its not clear what is causing the problem.

Optimization of a modularly-built campaign allows for the pieces to be taken apart, streamlined and put back together. Conversely, if an entire campaign is intertwined and locked tightly together, enhancing the whole is a very complex task.

A modular design allows thecampaign to launch in-market faster than acampaign designed as a single unit.

Single unit campaigns require all the content, landing pages, and emails be built before kick-off . Modularly-designed campaigns only need the initial logic flow to go to market. Additionally, if the flows are the same across buying cycles (or modules) copying them will save time.

Make each component of your marketing automation campaign a separate module. For example, each form, landing page, segment, filter, data store, field merges, dynamic content, etc. should be developed separately.

Units in a modularly-designed campaign interconnect and communicate, necessitating the use ofconsistentdata points throughout. For example, if a filter excludes one demographic by eliminating a certain value, other filters using that demographic should use theexact same value (even if the same result could be achieved a different way).

To create a campaign like a LEGO set, start at the lowest component level. Design first the data components, then the field merges, dynamic content, and forms that interface with those data components. Lastly, build out the landing pages, emails, and, finally, the campaign logic flows that use those units.

Test each unit as a single entity before incorporating it into the overall campaign. Do the same thing when the module is copied for reuse. That way you flush out any issues when its a single unit and not when there are multiple versions of it that you must chase down.

For example, when creating a landing page with a form, test that by itself first. Ensure the UI looks as expected in all browsers. Confirm the form functions as designed and that the data collected from the form saves correctly.

Once validated, test the landing page from the link in the email. Now, if something is not working, its easy to isolate the issue and resolve it beforeit has been propagated to multiple units that all need fixing.

Design each buying cycle of a campaign as a separate module. Use linkage to send contacts from one buying cycle to the next.

This makes the campaign logic easier to read when you open it back up in the future. At the same time,smaller modules are easier to reuse.

More importantly, by modularizing the buying cycle, it makes it easier to work in all the nuances of the business logic and to launch into the market sooner.

Implementing marketing automation as LEGO sets empowersyou for success, SixthDivision founderBrad Martineau explained in a podcast extolling this LEGO method. With a clear vision of the overall business need and associated strategy broken into individual components, campaigns become less overwhelming to build and simplerto put into the market.

Some opinions expressed in this article may be those of a guest author and not necessarily MarTech Today. Staff authors are listed here.

See the article here:

LEGO logic rocks marketing automation - MarTech Today

Machines are being used to automate a rising number of processes in the marketing world. Programmatic ad buying, artificial intelligence, chatbotsand other emerging technologies are changing the landscape and bringing with them new efficiencies and ways of working.

It followsthen, thatmarket research also stands to benefit from automation. One such example is the use of robots to test the effectiveness of advertising.

Brands want answers to specific creative campaign questions before their activity starts (around messaging and execution), while it is live (to assess how the target audience is responding) and when completed (to determine what ROIit has delivered).

Automation improves the accuracy of this kind of research, speeds up the collection of data and can save brands money. It also makes ad testing more intuitive and should mean human researchers can spend more time analysing the results to produce the actionable insight that brands crave.

READ MORE:Artificial intelligence: A force for good or evil?

One British multinational to invest in a fully-automated creative testing process is Reckitt Benckiser (RB), owner ofbrands including Dettol, Durex and Cillit Bang.

It is working with market research technology provider ZappiStore to improve ad testing results. Byautomating processes it has been able to increase the number of ads it tests in a year by 77%, from 188 to 333, which has resulted in a 14% jump in market effectiveness. Previously the business focused primarily on analysing the effectiveness ofTV commercials but it now examines creativity across Facebook, point of sale and online video.

Automation allows large brand owners to think and act with the agility of a startup.

The company has also increased the number of brands it can evaluate from 25 to 32, which has improved overall ad effectiveness from 45% to 59%.

Mathilde Levy, RBs senior consumer and market insight manager for Europe and the US, says automation has made a big difference.

In the FMCG world there is the internal clock of the marketing function and the external clock of the life of the customer. These two clocks are not always synchronised, she says. By using automation the internal clock is more in tune and allows large brand owners to think and act with the agility of a startup.

The company is keen to introduce automation into other areas of its research, such as qualitative work enablingmore interviews to becarried out in peoples homes using machines.

Automation is also removing barriers internally, says Levy. The research we provide is not always as timely as marketers want it to be and we can be regarded as blockers. Automation changes this. Instead of serving as a validation function and being the policemen, we become more of a consultant, providing insight that can help to drive the business forward.

RBis now looking to upgrade its automated efforts,which will enable its brand marketers to cross-compare previous studies with new ones to reveal trends that would previously have cost too much time and money to uncover.

For Dominic Grounsell, former global marketing director at Travelex and Marketing Society board member, it is the improved accuracy provided by automation which appeals.

I am not a fan of traditional quantitative and qualitative research methods and interviews on street corners that provide unnatural reactions when people are pressured for time, he says. Things have evolved, with tools such as neuroscience technology assessing the emotional impact of an ad or messaging.

He also points to the development of mobile phone tracking in providing more reliable analytics.

The more brands use [smartphones]to track consumer behaviour on a granular level the better answers they will get to more specific questions on where people go and how long they spend there.

Grounsell believestechnology that provides retailers with real-time location-based behavioural data about consumers, will be a huge help to marketers going forward.You can actually see people walking into a restaurant or a retailer, which provides a greater understanding of footfall, he says.

It is the ability to understandconsumerreactions in the momentthat prompted theNational Association for Stock Car Auto Racing (NASCAR) to beginconducting real-time fan research during races, enabling it to improve the experience for viewers and advertisers.

The organisation,which sanctions more than 1,200 races across the US, Canada, Mexico and Europe,understood thatpeople watch with their friends, chat about the races on social media and engaging with sponsorswhile drivers battle it out but neededto be part of the conversation in order to learn from it.

Working withmachine learning platform Remesh,NASCAR has been able to engage hundreds of thousands of fanssimultaneously to drive research.It works by a single moderator asking an open-ended question about a specific moment to the entire community but results in members of the fan council responding individually, providing qualitative data that NASCAR can then analyse.

People are complex, and to capture this complexity you need a human element.

Fans are asked for their views on which manufacturer has the most attractive car, what they think about the companies advertising and how they rate the race on a scale of one to10.

While some of the techniques outlined above are new, many brands have been using aspectsof automation in market research for years. Transport for London (TfL), for example, started using online quantitative methods to evaluate its different marketing campaigns 13 years ago.

TfL is now employing an element of automation in its qualitative work, via online discussion forums, and it has introduced mobile apps into its ethnographic research so customers can record their own experiences in real time.

TfLs customer insight manager Ian Pring welcomes the benefits of automation but says it is not time yet to lose the human element of market research completely.

TfL isnt just about running a transport system, we are a customer service business and were aware that people dont always report their behaviour, thoughts and feelings in a straightforward way, he says. People are complex, and to capture this complexity you need a human element, not just in qualitative fieldwork but in analysis, interpretation and integration of all customer insights.

READ MORE:Rise of the machinesAre robots after your job?

In the publishing sector Trinity Mirrors groups marketing director Zoe Harris says there is an internal debate on whether the automated analytics it uses for online brands would work for its print products.

Our editor-in-chief Lloyd Embley observes that if we used the same analysis for print as we do for online we would put the crossword on the front page, says Harris. Many people turn to itfirst and its the piece of content that people spend the most time engaged with.

There is a serious point to this observation. It demonstrates how in any content-driven business the data generated must inform long-term thinking and be authentic.

If all the research was automated it would not be as genuine. One of our most traditional and trusted methods of research is our postbag, says Harris.

Automation should be viewed as just another tool we can use as we try to do more with the same research budget.

Trinity Mirror also has its10,000-strong online research community Mouthpiece, which brings together readers and web visitors from more than 20 of its national and regional news brands. This provides feedback on how the big news stories are being covered and how people are feeling about the country or politics.

The research is informing our advertisers to help them get their content right, says Harris. We have seen, for instance, a shift from a north/south divide to more of a London versus the rest of the UK split. Brands need insight into underlying trends to remain in touch with mainstream Britain.

When it comes to automated ad testing, she does worry that it hinders creativity.

Some of the best ad campaigns are the ones that initially test badly. Automation should be viewed as just another tool we can use as we try to do more with the same research budget.

The speed of change in research automation is almost as rapid as NASCAR champion Kyle Busch flying around the Kentucky Speedway track at Julys Alsco 300. However, for the immediate future at least, robots and human researchers will co-exist so that brands benefit from the accuracy and pace of using computers without losing the creativity and gut feel of a real person.

See the article here:

The rise of automation in market research - Marketing Week

More efficient lighting and climate control is one of several cost-saving benefits a smart building can offer your customers.

Smart buildings are on the rise around the world as companies recognize the potential cost savings of automation.

With the advent of the Internet of Things (IoT), every building with an integrated security and access control system has the capability to also integrate the buildings energy use, water use, ventilation and more.

Here are five steps to ensure successful automation of a commercial facility.

The first step is to show your customers the potential savings by conducting a thorough cost analysis. When surveying, you should look at everything from air handlers and chillers, to irrigation and what types of lighting are currently in place.

Be sure to also include details like switching to LED lighting, and updating compressors and chillers, and show the total potential cost savings. Buildings waste a lot of energy.

Simply propping a door open can cause the automation system to go into overdrive, pumping out air and creating significant energy waste. The ROI on building automation can sometimes free up money for other projects, while enhancing technology, comfort and security.

Theres an ongoing shift in the decision-making authority from facility managers to the IT leadership. This can be problematic, as many CIOs are unfamiliar with the challenges of maintaining a facilitys physical security.

Facility managers can be equally frustrated by unfamiliar IoT technology. Helping to create or improve the relationship between the CIO and the facility manager is crucial.

Educating the CIO about physical security, and bridging the knowledge gaps for the facility manager with smart device technologies will be essential as the industry moves to a more IoT-centric mindset. Providers and integrators should present themselves as a coordination point.

As with any new technology, there will be those who are hesitant to embrace the change. For instance, the IT director may have concerns about putting all their eggs in one software basket.

When a building is fully automated and networked, a failure in one area can cause failure in others. There may also be network bandwidth limitations that will have to be addressed to handle the amount of data that will becollected and shared between the various components of an automated system.

Concerns can usually be alleviated through detailed communication and concrete information about how the buildings systems can live side-by-side and how integration can benefit the customer in the long run.

The most important thing is to ensure that everyone has an opportunity to voice those concerns at the beginning.

To ensure the long-term success of a project, its crucial that the technology specified is future-forward and will flex and grow as the institution does.

The transition to new technology is never seamless, but there are solutions that can simplify the process. Look for systems and devices that are open source in nature; meaning, they can work together or can be easily integrated with each other.

Choosing proprietary versus open systems and devices could limit overall functionality and long-term adaptability.

Transitioning to a fully optimized smart building may seem overwhelming at first. It can be helpful to outline a series of goals with timeline milestones set.

60 days: Set goals for improving efficiency. Compare energy costs across multiple locations to establish a baseline of use and areas of loss/waste.

Six months: Use the data from your test location to establish an initial ROI and look for areas to further improve.

12 months: Track energy use at all locations to measure energy savings and establish overall ROI.

24 months: Develop a long-term plan for energy savings, including how to integrate new facilities into the overall solution. Experienced players in the security space understand there are no shortcuts.

The adoption of IoT technologies and tools will initially grow at the pace at which industry experts can support it. Wait too long and the space will be disrupted by external forces driven by consumer demand.

Minu Youngkin is Marketing Manager at Allegion.

Read the rest here:

5 Steps to Successful Smart Building Automation - Security Sales & Integration