Why are we dubious about deep learning?

Katja Forbes

Katja Forbes founded syfte, a specialist business in research and experience design in 2014, and is an Australian pioneer in the field of experience design. Katja is an International Director on the Interaction Design Association Board. She is proud to be a co-founding member and Local Leader in Sydney of the global community organisation, Interaction Design Association. Together with Joe Ortenzi, Katja has built a community of over 1700 designers in Sydney, providing them with learning opportunities via lecture based meetups that draw a crowd of 150 people each time, a mentoring program and workshops.

 
The prospect of deep learning gives those of us in the industry something to get really excited about, and something to be nervous about, at the same time. 

If deep learning is artificial intelligence (AI) in its true form, it doesn’t require the interference or assistance of a human at all. This is truly revolutionary, but it can also feel a little precarious – we just can’t feel completely comfortable with learning that can be completely unsupervised. 

So should we do away with machine learning in our business, and the application of artificial intelligence that provides systems the ability to automatically carry out tasks, in favour of deep learning?  

So what is deep learning anyway? Well, think of it in relation to machine learning, which describes the process of teaching a computer to carry out a task. For example, this could be setting up your email signature so that it appears automatically each time you create, or replying to an email. Effectively, deep learning is a sub-field of machine learning, which is learning done unsupervised by a human.  

Believe it or not, deep learning has been around for quite a while, just not in the business or domestic domain, and already has a headstart in areas such as recognising speech or detecting cancer, opening up a whole range of possibilities. This gives us the idea that the computer has the ability to solve a whole host of problems that couldn’t otherwise be touched.

It is a positive result of course, but what does it mean for humanity when computers are learning things we don’t have a hope of understanding?     

Normally with machine learning, when there is a problem, we just jump in and reconfigure. With deep learning, chances are we have no idea what the computer is doing wrong, let alone how it is done right. A few options are then left to us, one of which is to destroy the faulty machine altogether, in case it begins to cause damage to its environment.   

Deep learning has been compared to the human mind, in that it can learn at the same level and speed at which the mind does, and solve much more complex problems than machine learning could ever tackle. Once again, we are delving into the stuff of Hollywood movies – will computers identified as deep learners solve problems that will enrich our lives in one way or another, or will they threaten the very fabric of our existence? Do we want to introduce something so sensitive into our work environment, something that can produce some extraordinary stuff, but also has the potential to go very wrong and destroy much more than it was even created in the first place? How would this influence our business, our clients, our livelihood? Are we brave enough to take that chance?  

Realistically, deep learning is not quite as advanced as the human mind. For example, deep learning requires a lot (as in thousands) of pictures of ‘people’ to work out how to recognise a person. So it can be easily fooled.

Will this current ‘problem’ be fixed? Most likely. The benefit is most of the work is done inside its ‘mind’ with little to no input required from a human, which is of great benefit to any business obviously until it starts to reach results that are undesired.  

At the moment, machine learning is doing the job perfectly 80 per cent of the time. This means that while you may wish it could achieve more, such as not require human-power in its learning, and perhaps some more precise results, it is likely not keeping you awake at night, something that deep learning may just do.  

Perhaps it’s better to allow deep learning to be understood better, ironed out, before introducing its remarkable ability into your workplace.  

 

 

 

 

 

 

 

 

 

 

Tags: Emerging Technologies, customer experience management, deep learning, leadership, artificial intelligence

Show Comments

Featured Whitepapers

State of the CMO 2019

CMO’s State of the CMO is an annual industry research initiative aimed at understanding how ...

More whitepapers

Blog Posts

AI ethics: Designing for trust

As artificial intelligence (AI) becomes much more prevalent and increasingly a way of life, more questions are being asked than answered about the ethical implications of its adoption.

Katja Forbes

Founder and chief, sfyte

How service heterogeneity is impacting engagement

Marketers have long known the importance of standardising products to assure quality and consistency. For services, however, standardisation is much more complex.

Dr Chris Baumann

Associate professor, Macquarie University

Kindness matters in business: why the nice guys finish first

A recent 1000-page Royal Commission report on misconduct in Australia’s financial sector revealed hair-raising stories of excessive commissions, rampant mis-selling and charges levied on the dead. So how do you stop a bank from misleading its customers?

Nick Liddell

Director of Consulting, The Clearing

nice article

meripadhai

5 things marketers need to do to get better in buy in when presenting

Read more

International business is closely related to marketing or marketing activities carried out by the company. According to Gitman and McDani...

Eko Prasetyo Utomo X

Salesforce: The age of the marketing campaign is over

Read more

Back in 1968 Holden began an appeal to customers who have an interest in competition. It did this with the introduction of the HK GTS 32...

Ben Tate

Marketing professor: For Holden, brand nostalgia ain’t what it used to be

Read more

Your blog post is really good and informative. Thanks for taking time to provide us this useful information with us.Auto wrapping uaeADF ...

Yes Machinery

Image intelligence:10 must-see infographics for marketers

Read more

A debt of gratitude is in order for sharing this marvelous information.I have taken in numerous things from your post

digitech Classes

Lumen CMO strives to make the brand synonymous with anti-ageism

Read more

Latest Podcast

More podcasts

Sign in