Player is loading...

Embed

Embed code

Transcriptions

Note: this content has been automatically generated.
00:00:00
right j. o. y.
00:00:08
oh
00:00:10
uh i just uttering actually i okay maybe that the minute actually address what
00:00:15
the claim is the following whatever say training das is the church irregular relation
00:00:21
or more secure so the are sort of like equivalent to each other so
00:00:25
there wouldn't yeah there was a gap year there was a small yeah yeah
00:00:32
i yeah but i guess you cannot beats what you get the adversarial training because the i guess this this one is
00:00:38
a kind of subset of that when you know what what seen it the other method das is sort of like you know
00:00:44
mm essentially i visit the training uh does good generalisation plus maybe something else
00:00:50
but maybe that's something as is not a uh uh
00:00:56
uh no the other bigger in city p. r. and that yeah but that's you know i i just brought some results yeah really have it
00:01:01
paper next to b. p. r. but shows that you know this kind of a good balance
00:01:12
look at it
00:01:17
for
00:01:24
i i
00:01:27
kind of transfer learning or to the bombing adaptation
00:01:37
well i
00:01:43
i'm not as a i haven't looked in that but
00:01:45
that's definitely an interesting question and pressing it being to discover
00:01:51
i i mailed is that because you know you you regular rise the model more i guess in two
00:01:56
senses maybe also you get some some use i i i don't i'm not sure that you get the bills
00:02:01
and then the for example the accuracy but maybe some something something else for example it might be a
00:02:06
more generalised able to kind of more data sets essentially maybe no you don't get a i like yeah that
00:02:13
the high performance but you know that number of data sets that
00:02:16
jen arises and you model would be more but that's that's my guess

Share this talk: 


Conference Program

Methods for Rule and Knowledge Extraction from Deep Neural Networks
Keynote speech: Prof. Pena Carlos Andrés, HEIG-VD
3 May 2019 · 9:10 a.m.
349 views
Q&A - Keynote speech: Prof. Pena Carlos Andrés
Keynote speech: Prof. Pena Carlos Andrés, HEIG-VD
3 May 2019 · 10:08 a.m.
Visualizing and understanding raw speech modeling with convolutional neural networks
Hannah Muckenhirn, Idiap Research Institute
3 May 2019 · 10:15 a.m.
Q&A - Hannah Muckenhirn
Hannah Muckenhirn, Idiap Research Institute
3 May 2019 · 10:28 a.m.
Concept Measures to Explain Deep Learning Predictions in Medical Imaging
Mara Graziani, HES-SO Valais-Wallis
3 May 2019 · 10:32 a.m.
What do neural network saliency maps encode?
Suraj Srinivas, Idiap Research Institute
3 May 2019 · 10:53 a.m.
Transparency of rotation-equivariant CNNs via local geometric priors
Dr Vincent Andrearczyk, HES-SO Valais-Wallis
3 May 2019 · 11:30 a.m.
Q&A - Dr Vincent Andrearczyk
Dr Vincent Andrearczyk, HES-SO Valais-Wallis
3 May 2019 · 11:48 a.m.
Interpretable models of robot motion learned from few demonstrations
Dr Sylvain Calinon, Idiap Research Institute
3 May 2019 · 11:50 a.m.
Q&A - Sylvain Calinon
Dr Sylvain Calinon, Idiap Research Institute
3 May 2019 · 12:06 p.m.
The HyperBagGraph DataEdron: An Enriched Browsing Experience of Scientific Publication Databa
Xavier Ouvrard, University of Geneva / CERN
3 May 2019 · 12:08 p.m.
Improving robustness to build more interpretable classifiers
Seyed Moosavi, Signal Processing Laboratory 4 (LTS4), EPFL
3 May 2019 · 12:21 p.m.
Q&A - Seyed Moosavi
Seyed Moosavi, Signal Processing Laboratory 4 (LTS4), EPFL
3 May 2019 · 12:34 p.m.