[CMU AI Seminar] Apr 6 at 12pm (Zoom) -- Elan Rosenfeld (CMU MLD) -- The Risks of Invariant Risk Minimization -- AI Seminar sponsored by Fortive

[Shaojie Bai]

Dear all,

We look forward to seeing you *next Tuesday (4/6)* from *1**2:00-1:00 PM
(U.S. Eastern time)* for the next talk of our *CMU AI seminar*, sponsored
by Fortive <https://careers.fortive.com/>.

To learn more about the seminar series or see the future schedule, please
visit the seminar website <http://www.cs.cmu.edu/~aiseminar/>.
<http://www.cs.cmu.edu/~aiseminar/>

On 4/6, *Elan Rosenfeld* (CMU MLD) will be giving a talk on "*The Risks of
Invariant Risk Minimization*."

*Title*: The Risks of Invariant Risk Minimization

*Talk Abstract*: Invariant feature learning has become a popular
alternative to Empirical Risk Minimization as practitioners recognize the
need to ignore features which may be misleading at test time in order to
improve out-of-distribution generalization. Early results in this area
leverage variation across environments to provably identify the features
which are directly causal with respect to the target variable. More recent
work attempts to use this technique for deep learning, frequently with no
formal guarantees of an algorithm's ability to uncover the correct
features. Most notably, the seminal work introducing Invariant Risk
Minimization gave a loose bound for the linear setting and no results at
all for non-linearity; despite this, a large number of variations have been
suggested. In this talk, I'll introduce a formal latent variable model
which encodes the primary assumptions made by these works. I'll then give
the first characterization of the optimal solution to the IRM objective,
deriving the exact number of environments needed for the solution to
generalize in the linear case. Finally, I'll present the first analysis of
IRM when the observed data is a non-linear function of the latent
variables: in particular, we show that IRM can fail catastrophically when
the test distribution is even moderately different from the training
distribution - this is exactly the problem that IRM was intended to solve.
These results easily generalize to all recent variations on IRM,
demonstrating that these works on invariant feature learning fundamentally
do not improve over standard ERM. This talk is based on work with Pradeep
Ravikumar and Andrej Risteski, to appear at ICLR 2021.

*Speaker Bio*: Elan Rosenfeld is a PhD student in the Machine Learning
Department at CMU, advised by Andrej Risteski and Pradeep Ravikumar. He is
interested in theoretical foundations of machine learning, with a
particular focus on robust learning, representation learning and
out-of-distribution generalization. Elan completed his undergraduate
degrees in Computer Science and Statistics & Machine Learning at CMU, where
his senior thesis on human-usable password schemas was advised by Manuel
Blum and Santosh Vempala.

*Zoom Link*:
https://cmu.zoom.us/j/96139997861?pwd=ZlMrUUZaWXY0Sm9mai9ZdjE4QXNDQT09


Thanks,
Shaojie Bai (MLD)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/ai-seminar-announce/attachments/20210330/3523507e/attachment.html>