Sherburne

  • Student

Submitted

“News”, “News”, “Fake News”: Can Machine Learning Help Identify Fake News on Facebook?
Sherburne
Posted on November 13, 2018 at 8:00 pm
Among those wishing for a do-over on the 2016 presidential election, Facebook is somewhere high on the list. Can machine learning help them avoid repeating their mistakes?

Activity Feed

On November 15, 2018, Sherburne commented on Soft robotics – How 3D printing enables technological advancements to explore oceans and save lives :

Interesting topic! Had not heard of soft robotics before now, but I can definitely see the potential.

To give a view on your question about whether to design for a niche application or a broader range, I think focusing on a specific field with many possible applications could give you the best of both worlds. For example, if ABB could create tiny “surgery robots” that could be used for many different types of surgeries, they could keep development costs low while maintaining a large potential market.

In general, the soft aspect of these robots makes them really promising for interaction with human beings (also soft). Of the interactions with the human body that require the greatest precision and pose the greatest risk of damage, surgery stood out to me, but there are clearly a ton of possibilities!

On November 15, 2018, Sherburne commented on Dangerous Innovation – Defense Distributed and the Democratization of Weapons Manufacturing :

I think this will be important not just for weapons, but also for products that–if not functionally sound–could injure consumers

On November 15, 2018, Sherburne commented on Dangerous Innovation – Defense Distributed and the Democratization of Weapons Manufacturing :

Appreciate you highlighting some of the risks that come along with the many exciting opportunities that additive manufacturing presents.

Ultimately, I believe that the manufacturers of 3D printers will have to collect and consolidate data about the items people print at home in order for any law enforcement body to manage these risks. The printers themselves seem like a more manageable point of observation than the many online databases of items that have already started to emerge, but in many countries–like the US–there will definitely be some privacy concerns.

I agree with @Christopher Reynolds that there is a risk of stifling innovation through regulation, but I think that governments have to know if individuals are manufacturing weapons. Safety should take priority over innovation.

On November 15, 2018, Sherburne commented on The government wants you…to hack it? :

Great illustration of a problem that is very well addressed by open innovation.

I think that extending the program to not only identify bugs but also to identify high-potential candidates for hiring would address some of the risks of the public “losing interest” and make the security improvements more sustainable. I think we saw this approach work quite well in the Valve case in our LEAD course, where outside video game “mod” developers were frequently hired full time.

I do agree that there are more security risks involved in government hiring relative to video game hiring, so additional background checking may be necessary.

On November 15, 2018, Sherburne commented on Open Innovation in CPG: The Future of HENRi@Nestlé :

Nice article about a cool initiative!

I imagine the relative success of courting marketing startups over technical startups has a lot to do with the incentives of partnering with Nestle over focusing on the startups’ own businesses. Whereas Nestle gives a marketing company something to market, a technology company may be more inclined to develop their own technology product and sell the product–or eventually the business–for a much greater payout than the $50,000 Nestle is offering.

If this is a big driver, Nestle may need to increase the offer for technical products in order to generate more participation.

On November 15, 2018, Sherburne commented on Increasing financial security with machine learning :

Very cool article and very cool company!

One question I had was whether you anticipate much conflict between the outcomes that employers are optimizing for and the ones that employees would optimize for with perfect information? Overall I agree with your thesis that there is a lot of alignment between good personal finance and good professional performance, but could there be some situations on the margin where the models would recommend solutions that are not in the best interests of the workers?

Particularly in the case of low-wage workers that you brought up, if the company is paying for this solution, and the suggestions are not truly mutually beneficial, who would you side with? One the one hand, you want to maximize the benefit to your customers (the companies), but you also want to avoid reputational or legal risks from hurting employees.

Maybe these will just be edge cases, but would be interested to hear more about it! Again, great idea, and good luck!

On November 14, 2018, Sherburne commented on Revolutionizing personal credit with machine learning :

Interesting topic and very accessible analysis of the competitive landscape for “subprime” lending. I have always felt like the stats that feed FICO scores seemed a bit arbitrary–a better measure of knowing what affects your score than being creditworthy–so it’s interesting to see some evidence that alternative metrics can be effective.

To your question about barriers to entry and customer offerings, I believe the main barrier to entry in financial services will continue to be building trust, but it may be that technology can help build that trust faster than the long track records that financial institution have needed in the past. I expect that the average person will become more literate in machine learning concepts in the near future, so marketing this approach will become easier, but also much more competitive. Ultimately it may be a single technology company that dominates the lending space with a truly revolutionary advancement in machine learning capability.