👨🏻‍💻 Mark E. Whiting

I'm a postdoc at Stanford under Michael Bernstein.

The difficulty of making good creative decisions constrains human productivity. My work focuses on how we might make creative decisions in the future, with help from computational systems and models of human behavior, improving speed and reliability.

My recent work studies four areas:

Automated design

To automate design, a computational representation of all the nuances of design is necessary. This has been challenging because even humans are usually unsure of the best ways to abstract design into a formal representation.

For my PhD, I have introduced an automatic, grammar-based system for structuring information (DCC'16). We applied this tool to data from design, architecture, and abstract systems (AI EDAM forthcoming). We used this representation technique for detecting subtle differences between data sets, as accurately as the existing state of the art, but requiring almost 10 times less data.

Crowd resourcing

Crowdsourcing tools today offer ad hoc and remote work arrangements at an unprecedented scale, yet many of the perks of traditional jobs have not yet been realized for this community; no benefits, limited support, and systems hindering economic mobility.

I am part of the Stanford Crowd Research Collective, a group of almost 1000 collaborators from around the world, led by Stanford professor Michael Bernstein. We conduct research as a crowd, and are developing Daemo (CSCW'17 demo), a new crowdsourcing platform that aims to mitigate many of the shortcomings of crowdsourcing today.

We also introduced crowd guilds (CSCW'17), a way to improve reputation and feedback within crowdsourcing, inspired by traditional worker guilds, and we have developed the Daemo Constitution (CI'17), a mechanism for self governance in crowdsourcing marketplaces.

Incentive design

Using incentives to motivate human behavior is very effective, but if the incentives are poorly designed the resulting behavior is rarely the desired one. For example, requiring a certain number of characters of feedback in an online feedback form is more likely to lead to feedback padded with spaces than higher quality responses, however this is a commonly used method in online settings. With algorithmic game theory, a branch of economics, incentives can be carefully designed to achieve specific behavioral goals.

Together with Dilrukshi Gamage, Thejan Rajapakshe, Haritha Thilakarathne, Indika Perera, and Shantha Fernando, we have explored how aligning the incentives for peer feedback in online learning helps improve the perceived quality and length of the feedback (L@S'17). We have used similar approaches for crowd guilds (CSCW'17) and boomerang (UIST'16) with the Stanford Crowd Research Collective.

Hybrid systems

Humans are pretty good at chess. Computers are better. Humans and computers together have been the best. We are still not very good at getting the best out of this kind of working relationship

Mentoring several groups of undergraduate students at CMU and in the Pitt i3 program, we have worked to build systems that are made stronger by encouraging humans and computers to work together. For example, We have explored how socio-technical systems can provide novel insight into new product development opportunities, by performing computational analysis on Amazon product reviews (iConference'16). In ongoing work, we are studying how novice humans can be supported by computer systems to provide rich design insight.