Hi all,
Our final virtual meeting of this semester will be at 12pm (EST) Wednesday,
December 2, where we will hear Kosuke Imai (Harvard University) presents
research on "Experimental Evaluation of Algorithm-Assisted Human
Decision-Making: Application to Pretrial Public Safety Assessment."
*Abstract*:
Despite an increasing reliance on fully-automated algorithmic decision
making in our day-to-day lives, human beings still make highly
consequential decisions. As frequently seen in business, healthcare, and
public policy, recommendations produced by algorithms are provided to human
decision-makers in order to guide their decisions. While there exists a
fast growing literature evaluating the bias and fairness of such
algorithmic recommendations, an overlooked question is whether they help
humans make better decisions. We develop a statistical methodology for
experimentally evaluating the causal impacts of algorithmic recommendations
on human decisions. We also show how to examine whether algorithmic
recommendations improve the fairness of human decisions and derive the
optimal decisions under various settings. We apply the proposed
methodology to the first-ever randomized controlled trial that evaluates
the pretrial public safety assessment (PSA) in the criminal justice system.
A goal of the PSA is to help judges decide which arrested individuals
should be released. We find that the PSA provision has little overall
impact on the judge’s decisions and subsequent arrestee behavior. However,
our analysis suggests that the PSA may help avoid unnecessarily harsh
decisions for female arrestees regardless of their risk levels while it
encourages the judge to make stricter decisions for male arrestees who are
deemed to be risky. In terms of fairness, the PSA appears to increase the
gender bias against males while having little effect on the existing racial
biases of the judge’s decisions against non-white males. Finally, we show
that PSA’s recommendations are often too severe and can only be justified
if the societal cost of a new crime is much higher than the cost of an
unnecessarily harsh decision.
*Zoom link*:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
*Schedule of the workshop*:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/
Hi all,
Our next virtual meeting will be at 12pm (EST) Wednesday, November 18,
where we will hear Tyler VanderWeele (Harvard University) presents research
on "Revisiting Psychometric Theory and Factor Analysis."
*Abstract*:
The presentation will revisit some of the conceptual and statistical
foundations of psychometric measurement theory and factor analysis,
specifically addressing the questions: (i) What happens to “factors” when
they causally affect one another?, (ii) Is an underlying univariate latent
variable a reasonable model for psycho-social constructs?, (iii) What are
the testable empirical implications of such a model? and (iv) What
alternative interpretations of analyses with constructed measures might be
possible?
The presentation will be based upon the following three preprints:
- VanderWeele, T.J. and Batty, C.J.K. (2020). On the dimensional
indeterminacy of one-wave factor analysis under causal effects.
<https://arxiv.org/abs/2001.10352> Technical Report.
- VanderWeele, T.J. and Vansteelandt, S. (2020). A statistical test to
reject the structural interpretation of a latent factor model
<https://arxiv.org/abs/2006.15899>. Technical Report.
- VanderWeele, T.J. (2020). Causal inference and constructed measures:
towards a new model of measurement for psychosocial constructs.
<https://arxiv.org/abs/2007.00520> Technical Report.
*Zoom link*:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
*Schedule of the workshop*:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/
Hi all,
Our next virtual meeting will be at 12pm (EST) Wednesday, November 11,
where we will hear Cory McCartan (Harvard University) presents research on
"Sequential Monte Carlo for Sampling Balanced and Compact Redistricting
Plans."
*Abstract*:
Random sampling of graph partitions under constraints has become a popular
tool for evaluating legislative redistricting plans. Analysts detect
partisan gerrymandering by comparing a proposed redistricting plan with an
ensemble of sampled alternative plans. For successful application, sampling
methods must scale to large maps with many districts, incorporate realistic
legal constraints, and accurately sample from a selected target
distribution. Unfortunately, most existing methods struggle in at least one
of these three areas. We present a new Sequential Monte Carlo (SMC)
algorithm that draws representative redistricting plans from a realistic
target distribution of choice. Because it yields nearly independent
samples, the SMC algorithm can efficiently explore the relevant space of
redistricting plans than the existing Markov chain Monte Carlo algorithms
that yield dependent samples. Our algorithm can simultaneously incorporate
several constraints commonly imposed in real-world redistricting problems,
including equal population, compactness, and preservation of administrative
boundaries. We validate the accuracy of the proposed algorithm by using a
small map where all redistricting plans can be enumerated. We then apply
the SMC algorithm to evaluate the partisan implications of several maps
submitted by relevant parties in a recent high-profile redistricting case
in the state of Pennsylvania. Open-source software is available for
implementing the proposed methodology.
Paper is available from here <https://arxiv.org/pdf/2008.06131.pdf>.
*Zoom link*:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
*Schedule of the workshop*:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/
Hi all,
Our next virtual meeting will be at 12pm (EST) Wednesday, November 4,
where we will hear Yiling Chen (Harvard University) presents research on
"Unexpected Consequences of Algorithm-in-the-Loop Decision Making."
*Abstract*:
The rise of machine learning has fundamentally altered decision making:
rather than being made solely by people, many important decisions are now
made through an “algorithm-in-the-loop” process where machine learning
models inform people. Yet insufficient research has considered how the
interactions between people and models actually influence human decision
making. In this talk, I’ll discuss results from a set of controlled
experiments on algorithm-in-the-loop human decision making in two contexts
(pretrial release and financial lending). For example, when presented with
algorithmic risk assessments, our study participants exhibited additional
bias in their decisions and showed a change in their decision-making
process by increasing risk aversion. These results highlight the urgent
need to expand our analyses of algorithmic decision making aids beyond
evaluating the models themselves to investigating the full sociotechnical
contexts in which people and algorithms interact.
Papers of the talk are available from here
<https://scholar.harvard.edu/files/bgreen/files/19-cscw.pdf> and here
<https://scholar.harvard.edu/files/19-fat.pdf>.
*Zoom link*:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
*Schedule of the workshop*:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/