---
OA_place: repository
OA_type: green
_id: '10597'
abstract:
- lang: eng
text: We thank Emmanuel Abbe and Min Ye for providing us the implementation of RPA
decoding. D. Fathollahi and M. Mondelli are partially supported by the 2019 Lopez-Loreta
Prize. N. Farsad is supported by Discovery Grant from the Natural Sciences and
Engineering Research Council of Canada (NSERC) and Canada Foundation for Innovation
(CFI), John R. Evans Leader Fund. S. A. Hashemi is supported by a Postdoctoral
Fellowship from NSERC.
article_processing_charge: No
arxiv: 1
author:
- first_name: Dorsa
full_name: Fathollahi, Dorsa
id: 712472af-8a7e-11f1-848f-9768b0c2fdf8
last_name: Fathollahi
- first_name: Nariman
full_name: Farsad, Nariman
last_name: Farsad
- first_name: Seyyed Ali
full_name: Hashemi, Seyyed Ali
last_name: Hashemi
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
citation:
ama: 'Fathollahi D, Farsad N, Hashemi SA, Mondelli M. Sparse multi-decoder recursive
projection aggregation for Reed-Muller codes. In: 2021 IEEE International Symposium
on Information Theory. IEEE; 2021:1082-1087. doi:10.1109/isit45174.2021.9517887'
apa: 'Fathollahi, D., Farsad, N., Hashemi, S. A., & Mondelli, M. (2021). Sparse
multi-decoder recursive projection aggregation for Reed-Muller codes. In 2021
IEEE International Symposium on Information Theory (pp. 1082–1087). Virtual,
Melbourne, Australia: IEEE. https://doi.org/10.1109/isit45174.2021.9517887'
chicago: Fathollahi, Dorsa, Nariman Farsad, Seyyed Ali Hashemi, and Marco Mondelli.
“Sparse Multi-Decoder Recursive Projection Aggregation for Reed-Muller Codes.”
In 2021 IEEE International Symposium on Information Theory, 1082–87. IEEE,
2021. https://doi.org/10.1109/isit45174.2021.9517887.
ieee: D. Fathollahi, N. Farsad, S. A. Hashemi, and M. Mondelli, “Sparse multi-decoder
recursive projection aggregation for Reed-Muller codes,” in 2021 IEEE International
Symposium on Information Theory, Virtual, Melbourne, Australia, 2021, pp.
1082–1087.
ista: 'Fathollahi D, Farsad N, Hashemi SA, Mondelli M. 2021. Sparse multi-decoder
recursive projection aggregation for Reed-Muller codes. 2021 IEEE International
Symposium on Information Theory. ISIT: International Symposium on Information
Theory, 1082–1087.'
mla: Fathollahi, Dorsa, et al. “Sparse Multi-Decoder Recursive Projection Aggregation
for Reed-Muller Codes.” 2021 IEEE International Symposium on Information Theory,
IEEE, 2021, pp. 1082–87, doi:10.1109/isit45174.2021.9517887.
short: D. Fathollahi, N. Farsad, S.A. Hashemi, M. Mondelli, in:, 2021 IEEE International
Symposium on Information Theory, IEEE, 2021, pp. 1082–1087.
conference:
end_date: 2021-07-20
location: Virtual, Melbourne, Australia
name: 'ISIT: International Symposium on Information Theory'
start_date: 2021-07-12
date_created: 2022-01-03T11:31:26Z
date_published: 2021-09-01T00:00:00Z
date_updated: 2026-08-12T06:36:19Z
day: '01'
department:
- _id: MaMo
doi: 10.1109/isit45174.2021.9517887
external_id:
arxiv:
- '2011.12882'
isi:
- '000701502201029'
fulldoi: https://doi.org/10.1109/isit45174.2021.9517887
isi: 1
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2011.12882
month: '09'
oa: 1
oa_version: Preprint
page: 1082-1087
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: 2021 IEEE International Symposium on Information Theory
publication_identifier:
eisbn:
- 978-1-5386-8209-8
isbn:
- 978-1-5386-8210-4
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Sparse multi-decoder recursive projection aggregation for Reed-Muller codes
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
year: '2021'
...
---
_id: '10053'
abstract:
- lang: eng
text: 'This paper characterizes the latency of the simplified successive-cancellation
(SSC) decoding scheme for polar codes under hardware resource constraints. In
particular, when the number of processing elements P that can perform SSC decoding
operations in parallel is limited, as is the case in practice, the latency of
SSC decoding is O(N1−1 μ+NPlog2log2NP), where N is the block length of the code
and μ is the scaling exponent of polar codes for the channel. Three direct consequences
of this bound are presented. First, in a fully-parallel implementation where P=N2
, the latency of SSC decoding is O(N1−1/μ) , which is sublinear in the block length.
This recovers a result from an earlier work. Second, in a fully-serial implementation
where P=1 , the latency of SSC decoding scales as O(Nlog2log2N) . The multiplicative
constant is also calculated: we show that the latency of SSC decoding when P=1
is given by (2+o(1))Nlog2log2N . Third, in a semi-parallel implementation, the
smallest P that gives the same latency as that of the fully-parallel implementation
is P=N1/μ . The tightness of our bound on SSC decoding latency and the applicability
of the foregoing results is validated through extensive simulations.'
acknowledgement: "S. A. Hashemi is supported by a Postdoctoral Fellowship from the
Natural Sciences and Engineering Research Council\r\nof Canada (NSERC) and by Huawei.
M. Mondelli is partially supported by the 2019 Lopez-Loreta Prize. A. Fazeli and
A. Vardy were supported in part by the National Science Foundation under Grant CCF-1764104."
article_processing_charge: No
arxiv: 1
author:
- first_name: Seyyed Ali
full_name: Hashemi, Seyyed Ali
last_name: Hashemi
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Arman
full_name: Fazeli, Arman
last_name: Fazeli
- first_name: Alexander
full_name: Vardy, Alexander
last_name: Vardy
- first_name: John
full_name: Cioffi, John
last_name: Cioffi
- first_name: Andrea
full_name: Goldsmith, Andrea
last_name: Goldsmith
citation:
ama: 'Hashemi SA, Mondelli M, Fazeli A, Vardy A, Cioffi J, Goldsmith A. Parallelism
versus latency in simplified successive-cancellation decoding of polar codes.
In: 2021 IEEE International Symposium on Information Theory. IEEE; 2021:2369-2374.
doi:10.1109/ISIT45174.2021.9518153'
apa: 'Hashemi, S. A., Mondelli, M., Fazeli, A., Vardy, A., Cioffi, J., & Goldsmith,
A. (2021). Parallelism versus latency in simplified successive-cancellation decoding
of polar codes. In 2021 IEEE International Symposium on Information Theory
(pp. 2369–2374). Melbourne, Australia: IEEE. https://doi.org/10.1109/ISIT45174.2021.9518153'
chicago: Hashemi, Seyyed Ali, Marco Mondelli, Arman Fazeli, Alexander Vardy, John
Cioffi, and Andrea Goldsmith. “Parallelism versus Latency in Simplified Successive-Cancellation
Decoding of Polar Codes.” In 2021 IEEE International Symposium on Information
Theory, 2369–74. IEEE, 2021. https://doi.org/10.1109/ISIT45174.2021.9518153.
ieee: S. A. Hashemi, M. Mondelli, A. Fazeli, A. Vardy, J. Cioffi, and A. Goldsmith,
“Parallelism versus latency in simplified successive-cancellation decoding of
polar codes,” in 2021 IEEE International Symposium on Information Theory,
Melbourne, Australia, 2021, pp. 2369–2374.
ista: 'Hashemi SA, Mondelli M, Fazeli A, Vardy A, Cioffi J, Goldsmith A. 2021. Parallelism
versus latency in simplified successive-cancellation decoding of polar codes.
2021 IEEE International Symposium on Information Theory. ISIT: International Symposium
on Information Theory, 2369–2374.'
mla: Hashemi, Seyyed Ali, et al. “Parallelism versus Latency in Simplified Successive-Cancellation
Decoding of Polar Codes.” 2021 IEEE International Symposium on Information
Theory, IEEE, 2021, pp. 2369–74, doi:10.1109/ISIT45174.2021.9518153.
short: S.A. Hashemi, M. Mondelli, A. Fazeli, A. Vardy, J. Cioffi, A. Goldsmith,
in:, 2021 IEEE International Symposium on Information Theory, IEEE, 2021, pp.
2369–2374.
conference:
end_date: 2021-07-20
location: Melbourne, Australia
name: 'ISIT: International Symposium on Information Theory'
start_date: 2021-07-12
date_created: 2021-09-27T14:33:14Z
date_published: 2021-09-01T00:00:00Z
date_updated: 2026-08-12T06:43:01Z
day: '01'
department:
- _id: MaMo
doi: 10.1109/ISIT45174.2021.9518153
external_id:
arxiv:
- '2012.13378'
isi:
- '000701502202078'
fulldoi: https://doi.org/10.1109/ISIT45174.2021.9518153
isi: 1
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2012.13378
month: '09'
oa: 1
oa_version: Preprint
page: 2369-2374
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: 2021 IEEE International Symposium on Information Theory
publication_identifier:
eisbn:
- 978-1-5386-8209-8
isbn:
- 978-1-5386-8210-4
issn:
- 2157-8095
publication_status: published
publisher: IEEE
quality_controlled: '1'
related_material:
record:
- id: '10364'
relation: later_version
status: public
scopus_import: '1'
status: public
title: Parallelism versus latency in simplified successive-cancellation decoding of
polar codes
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
year: '2021'
...
---
_id: '9047'
abstract:
- lang: eng
text: This work analyzes the latency of the simplified successive cancellation (SSC)
decoding scheme for polar codes proposed by Alamdar-Yazdi and Kschischang. It
is shown that, unlike conventional successive cancellation decoding, where latency
is linear in the block length, the latency of SSC decoding is sublinear. More
specifically, the latency of SSC decoding is O(N1−1/μ) , where N is the block
length and μ is the scaling exponent of the channel, which captures the speed
of convergence of the rate to capacity. Numerical results demonstrate the tightness
of the bound and show that most of the latency reduction arises from the parallel
decoding of subcodes of rate 0 or 1.
acknowledgement: M. Mondelli was partially supported by grants NSF DMS-1613091, CCF-1714305,
IIS-1741162, and ONR N00014-18-1-2729. S. A. Hashemi is supported by a Postdoctoral
Fellowship from the Natural Sciences and Engineering Research Council of Canada
(NSERC) and by Huawei. The authors would like to thank the anonymous reviewers for
their comments that helped improving the quality of the manuscript.
article_processing_charge: No
article_type: original
arxiv: 1
author:
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Seyyed Ali
full_name: Hashemi, Seyyed Ali
last_name: Hashemi
- first_name: John M.
full_name: Cioffi, John M.
last_name: Cioffi
- first_name: Andrea
full_name: Goldsmith, Andrea
last_name: Goldsmith
citation:
ama: Mondelli M, Hashemi SA, Cioffi JM, Goldsmith A. Sublinear latency for simplified
successive cancellation decoding of polar codes. IEEE Transactions on Wireless
Communications. 2021;20(1):18-27. doi:10.1109/TWC.2020.3022922
apa: Mondelli, M., Hashemi, S. A., Cioffi, J. M., & Goldsmith, A. (2021). Sublinear
latency for simplified successive cancellation decoding of polar codes. IEEE
Transactions on Wireless Communications. IEEE. https://doi.org/10.1109/TWC.2020.3022922
chicago: Mondelli, Marco, Seyyed Ali Hashemi, John M. Cioffi, and Andrea Goldsmith.
“Sublinear Latency for Simplified Successive Cancellation Decoding of Polar Codes.”
IEEE Transactions on Wireless Communications. IEEE, 2021. https://doi.org/10.1109/TWC.2020.3022922.
ieee: M. Mondelli, S. A. Hashemi, J. M. Cioffi, and A. Goldsmith, “Sublinear latency
for simplified successive cancellation decoding of polar codes,” IEEE Transactions
on Wireless Communications, vol. 20, no. 1. IEEE, pp. 18–27, 2021.
ista: Mondelli M, Hashemi SA, Cioffi JM, Goldsmith A. 2021. Sublinear latency for
simplified successive cancellation decoding of polar codes. IEEE Transactions
on Wireless Communications. 20(1), 18–27.
mla: Mondelli, Marco, et al. “Sublinear Latency for Simplified Successive Cancellation
Decoding of Polar Codes.” IEEE Transactions on Wireless Communications,
vol. 20, no. 1, IEEE, 2021, pp. 18–27, doi:10.1109/TWC.2020.3022922.
short: M. Mondelli, S.A. Hashemi, J.M. Cioffi, A. Goldsmith, IEEE Transactions on
Wireless Communications 20 (2021) 18–27.
corr_author: '1'
date_created: 2021-01-31T23:01:21Z
date_published: 2021-01-01T00:00:00Z
date_updated: 2026-08-12T11:12:22Z
day: '01'
department:
- _id: MaMo
doi: 10.1109/TWC.2020.3022922
external_id:
arxiv:
- '1909.04892'
isi:
- '000607808800002'
fulldoi: https://doi.org/10.1109/TWC.2020.3022922
intvolume: ' 20'
isi: 1
issue: '1'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/1909.04892
month: '01'
oa: 1
oa_version: Preprint
page: 18-27
publication: IEEE Transactions on Wireless Communications
publication_identifier:
eissn:
- 1558-2248
issn:
- 1536-1276
publication_status: published
publisher: IEEE
quality_controlled: '1'
related_material:
record:
- id: '8536'
relation: earlier_version
status: public
scopus_import: '1'
status: public
title: Sublinear latency for simplified successive cancellation decoding of polar
codes
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 20
year: '2021'
...
---
_id: '9221'
abstract:
- lang: eng
text: "Recent works have shown that gradient descent can find a global minimum for
over-parameterized neural networks where the widths of all the hidden layers scale
polynomially with N (N being the number of training samples). In this paper, we
prove that, for deep networks, a single layer of width N following the input layer
suffices to ensure a similar guarantee. In particular, all the remaining layers
are allowed to have constant widths, and form a pyramidal topology. We show an
application of our result to the widely used LeCun’s initialization and obtain
an over-parameterization requirement for the single wide layer of order N2.\r\n"
acknowledgement: The authors would like to thank Jan Maas, Mahdi Soltanolkotabi, and
Daniel Soudry for the helpful discussions, Marius Kloft, Matthias Hein and Quoc
Dinh Tran for proofreading portions of a prior version of this paper, and James
Martens for a clarification concerning LeCun’s initialization. M. Mondelli was partially
supported by the 2019 Lopez-Loreta Prize. Q. Nguyen was partially supported by the
German Research Foundation (DFG) award KL 2698/2-1.
alternative_title:
- Advances in Neural Information Processing Systems
article_processing_charge: No
arxiv: 1
author:
- first_name: Quynh
full_name: Nguyen, Quynh
last_name: Nguyen
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
citation:
ama: 'Nguyen Q, Mondelli M. Global convergence of deep networks with one wide layer
followed by pyramidal topology. In: 34th Conference on Neural Information Processing
Systems. Vol 33. Neural Information Processing Systems Foundation; 2020:11961–11972.'
apa: 'Nguyen, Q., & Mondelli, M. (2020). Global convergence of deep networks
with one wide layer followed by pyramidal topology. In 34th Conference on Neural
Information Processing Systems (Vol. 33, pp. 11961–11972). Vancouver, Canada:
Neural Information Processing Systems Foundation.'
chicago: Nguyen, Quynh, and Marco Mondelli. “Global Convergence of Deep Networks
with One Wide Layer Followed by Pyramidal Topology.” In 34th Conference on
Neural Information Processing Systems, 33:11961–11972. Neural Information
Processing Systems Foundation, 2020.
ieee: Q. Nguyen and M. Mondelli, “Global convergence of deep networks with one wide
layer followed by pyramidal topology,” in 34th Conference on Neural Information
Processing Systems, Vancouver, Canada, 2020, vol. 33, pp. 11961–11972.
ista: 'Nguyen Q, Mondelli M. 2020. Global convergence of deep networks with one
wide layer followed by pyramidal topology. 34th Conference on Neural Information
Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in
Neural Information Processing Systems, vol. 33, 11961–11972.'
mla: Nguyen, Quynh, and Marco Mondelli. “Global Convergence of Deep Networks with
One Wide Layer Followed by Pyramidal Topology.” 34th Conference on Neural Information
Processing Systems, vol. 33, Neural Information Processing Systems Foundation,
2020, pp. 11961–11972.
short: Q. Nguyen, M. Mondelli, in:, 34th Conference on Neural Information Processing
Systems, Neural Information Processing Systems Foundation, 2020, pp. 11961–11972.
conference:
end_date: 2020-12-12
location: Vancouver, Canada
name: 'NeurIPS: Neural Information Processing Systems'
start_date: 2020-12-06
date_created: 2021-03-03T12:06:02Z
date_published: 2020-07-07T00:00:00Z
date_updated: 2025-05-14T11:27:13Z
day: '07'
department:
- _id: MaMo
external_id:
arxiv:
- '2002.07867'
intvolume: ' 33'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2002.07867
month: '07'
oa: 1
oa_version: Preprint
page: 11961–11972
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: 34th Conference on Neural Information Processing Systems
publication_status: published
publisher: Neural Information Processing Systems Foundation
quality_controlled: '1'
status: public
title: Global convergence of deep networks with one wide layer followed by pyramidal
topology
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 33
year: '2020'
...
---
_id: '6748'
abstract:
- lang: eng
text: "Fitting a function by using linear combinations of a large number N of `simple'
components is one of the most fruitful ideas in statistical learning. This idea
lies at the core of a variety of methods, from two-layer neural networks to kernel
regression, to boosting. In general, the resulting risk minimization problem is
non-convex and is solved by gradient descent or its variants. Unfortunately, little
is known about global convergence properties of these approaches.\r\nHere we consider
the problem of learning a concave function f on a compact convex domain Ω⊆ℝd,
using linear combinations of `bump-like' components (neurons). The parameters
to be fitted are the centers of N bumps, and the resulting empirical risk minimization
problem is highly non-convex. We prove that, in the limit in which the number
of neurons diverges, the evolution of gradient descent converges to a Wasserstein
gradient flow in the space of probability distributions over Ω. Further, when
the bump width δ tends to 0, this gradient flow has a limit which is a viscous
porous medium equation. Remarkably, the cost function optimized by this gradient
flow exhibits a special property known as displacement convexity, which implies
exponential convergence rates for N→∞, δ→0. Surprisingly, this asymptotic theory
appears to capture well the behavior for moderate values of δ,N. Explaining this
phenomenon, and understanding the dependence on δ,N in a quantitative manner remains
an outstanding challenge."
article_processing_charge: No
article_type: original
arxiv: 1
author:
- first_name: Adel
full_name: Javanmard, Adel
last_name: Javanmard
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Andrea
full_name: Montanari, Andrea
last_name: Montanari
citation:
ama: Javanmard A, Mondelli M, Montanari A. Analysis of a two-layer neural network
via displacement convexity. Annals of Statistics. 2020;48(6):3619-3642.
doi:10.1214/20-AOS1945
apa: Javanmard, A., Mondelli, M., & Montanari, A. (2020). Analysis of a two-layer
neural network via displacement convexity. Annals of Statistics. Institute
of Mathematical Statistics. https://doi.org/10.1214/20-AOS1945
chicago: Javanmard, Adel, Marco Mondelli, and Andrea Montanari. “Analysis of a Two-Layer
Neural Network via Displacement Convexity.” Annals of Statistics. Institute
of Mathematical Statistics, 2020. https://doi.org/10.1214/20-AOS1945.
ieee: A. Javanmard, M. Mondelli, and A. Montanari, “Analysis of a two-layer neural
network via displacement convexity,” Annals of Statistics, vol. 48, no.
6. Institute of Mathematical Statistics, pp. 3619–3642, 2020.
ista: Javanmard A, Mondelli M, Montanari A. 2020. Analysis of a two-layer neural
network via displacement convexity. Annals of Statistics. 48(6), 3619–3642.
mla: Javanmard, Adel, et al. “Analysis of a Two-Layer Neural Network via Displacement
Convexity.” Annals of Statistics, vol. 48, no. 6, Institute of Mathematical
Statistics, 2020, pp. 3619–42, doi:10.1214/20-AOS1945.
short: A. Javanmard, M. Mondelli, A. Montanari, Annals of Statistics 48 (2020) 3619–3642.
date_created: 2019-07-31T09:39:42Z
date_published: 2020-12-11T00:00:00Z
date_updated: 2024-10-21T06:02:33Z
day: '11'
department:
- _id: MaMo
doi: 10.1214/20-AOS1945
external_id:
arxiv:
- '1901.01375'
isi:
- '000598369200021'
fulldoi: https://doi.org/10.1214/20-AOS1945
intvolume: ' 48'
isi: 1
issue: '6'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/1901.01375
month: '12'
oa: 1
oa_version: Preprint
page: 3619-3642
publication: Annals of Statistics
publication_identifier:
eissn:
- 1941-7330
issn:
- 1932-6157
publication_status: published
publisher: Institute of Mathematical Statistics
quality_controlled: '1'
scopus_import: '1'
status: public
title: Analysis of a two-layer neural network via displacement convexity
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 48
year: '2020'
...
---
_id: '8536'
abstract:
- lang: eng
text: This work analyzes the latency of the simplified successive cancellation (SSC)
decoding scheme for polar codes proposed by Alamdar-Yazdi and Kschischang. It
is shown that, unlike conventional successive cancellation decoding, where latency
is linear in the block length, the latency of SSC decoding is sublinear. More
specifically, the latency of SSC decoding is O(N 1−1/µ ), where N is the block
length and µ is the scaling exponent of the channel, which captures the speed
of convergence of the rate to capacity. Numerical results demonstrate the tightness
of the bound and show that most of the latency reduction arises from the parallel
decoding of subcodes of rate 0 and 1.
acknowledgement: M. Mondelli was partially supported by grants NSF DMS-1613091, CCF-1714305,
IIS-1741162 and ONR N00014-18-1-2729. S. A. Hashemi is supported by a Postdoctoral
Fellowship from the Natural Sciences and Engineering Research Council of Canada
(NSERC) and by Huawei.
article_number: 401-406
article_processing_charge: No
arxiv: 1
author:
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Seyyed Ali
full_name: Hashemi, Seyyed Ali
last_name: Hashemi
- first_name: John
full_name: Cioffi, John
last_name: Cioffi
- first_name: Andrea
full_name: Goldsmith, Andrea
last_name: Goldsmith
citation:
ama: 'Mondelli M, Hashemi SA, Cioffi J, Goldsmith A. Simplified successive cancellation
decoding of polar codes has sublinear latency. In: IEEE International Symposium
on Information Theory. Vol 2020-June. IEEE; 2020. doi:10.1109/ISIT44484.2020.9174141'
apa: 'Mondelli, M., Hashemi, S. A., Cioffi, J., & Goldsmith, A. (2020). Simplified
successive cancellation decoding of polar codes has sublinear latency. In IEEE
International Symposium on Information Theory (Vol. 2020–June). Los Angeles,
CA, United States: IEEE. https://doi.org/10.1109/ISIT44484.2020.9174141'
chicago: Mondelli, Marco, Seyyed Ali Hashemi, John Cioffi, and Andrea Goldsmith.
“Simplified Successive Cancellation Decoding of Polar Codes Has Sublinear Latency.”
In IEEE International Symposium on Information Theory, Vol. 2020–June.
IEEE, 2020. https://doi.org/10.1109/ISIT44484.2020.9174141.
ieee: M. Mondelli, S. A. Hashemi, J. Cioffi, and A. Goldsmith, “Simplified successive
cancellation decoding of polar codes has sublinear latency,” in IEEE International
Symposium on Information Theory, Los Angeles, CA, United States, 2020, vol.
2020–June.
ista: 'Mondelli M, Hashemi SA, Cioffi J, Goldsmith A. 2020. Simplified successive
cancellation decoding of polar codes has sublinear latency. IEEE International
Symposium on Information Theory. ISIT: International Symposium on Information
Theory vol. 2020–June, 401–406.'
mla: Mondelli, Marco, et al. “Simplified Successive Cancellation Decoding of Polar
Codes Has Sublinear Latency.” IEEE International Symposium on Information Theory,
vol. 2020–June, 401–406, IEEE, 2020, doi:10.1109/ISIT44484.2020.9174141.
short: M. Mondelli, S.A. Hashemi, J. Cioffi, A. Goldsmith, in:, IEEE International
Symposium on Information Theory, IEEE, 2020.
conference:
end_date: 2020-06-26
location: Los Angeles, CA, United States
name: 'ISIT: International Symposium on Information Theory'
start_date: 2020-06-21
date_created: 2020-09-20T22:01:37Z
date_published: 2020-06-01T00:00:00Z
date_updated: 2026-08-12T11:12:23Z
day: '01'
department:
- _id: MaMo
doi: 10.1109/ISIT44484.2020.9174141
external_id:
arxiv:
- '1909.04892'
isi:
- '000714963400069'
fulldoi: https://doi.org/10.1109/ISIT44484.2020.9174141
isi: 1
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/1909.04892
month: '06'
oa: 1
oa_version: Preprint
publication: IEEE International Symposium on Information Theory
publication_identifier:
isbn:
- '9781728164328'
issn:
- 2157-8095
publication_status: published
publisher: IEEE
quality_controlled: '1'
related_material:
record:
- id: '9047'
relation: later_version
status: public
scopus_import: '1'
status: public
title: Simplified successive cancellation decoding of polar codes has sublinear latency
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 2020-June
year: '2020'
...
---
_id: '9198'
abstract:
- lang: eng
text: "The optimization of multilayer neural networks typically leads to a solution\r\nwith
zero training error, yet the landscape can exhibit spurious local minima\r\nand
the minima can be disconnected. In this paper, we shed light on this\r\nphenomenon:
we show that the combination of stochastic gradient descent (SGD)\r\nand over-parameterization
makes the landscape of multilayer neural networks\r\napproximately connected and
thus more favorable to optimization. More\r\nspecifically, we prove that SGD solutions
are connected via a piecewise linear\r\npath, and the increase in loss along this
path vanishes as the number of\r\nneurons grows large. This result is a consequence
of the fact that the\r\nparameters found by SGD are increasingly dropout stable
as the network becomes\r\nwider. We show that, if we remove part of the neurons
(and suitably rescale the\r\nremaining ones), the change in loss is independent
of the total number of\r\nneurons, and it depends only on how many neurons are
left. Our results exhibit\r\na mild dependence on the input dimension: they are
dimension-free for two-layer\r\nnetworks and depend linearly on the dimension
for multilayer networks. We\r\nvalidate our theoretical findings with numerical
experiments for different\r\narchitectures and classification tasks."
acknowledgement: M. Mondelli was partially supported by the 2019 LopezLoreta Prize.
The authors thank Phan-Minh Nguyen for helpful discussions and the IST Distributed
Algorithms and Systems Lab for providing computational resources.
article_processing_charge: No
arxiv: 1
author:
- first_name: Aleksandr
full_name: Shevchenko, Aleksandr
id: F2B06EC2-C99E-11E9-89F0-752EE6697425
last_name: Shevchenko
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
citation:
ama: 'Shevchenko A, Mondelli M. Landscape connectivity and dropout stability of
SGD solutions for over-parameterized neural networks. In: Proceedings of the
37th International Conference on Machine Learning. Vol 119. ML Research Press;
2020:8773-8784.'
apa: Shevchenko, A., & Mondelli, M. (2020). Landscape connectivity and dropout
stability of SGD solutions for over-parameterized neural networks. In Proceedings
of the 37th International Conference on Machine Learning (Vol. 119, pp. 8773–8784).
ML Research Press.
chicago: Shevchenko, Aleksandr, and Marco Mondelli. “Landscape Connectivity and
Dropout Stability of SGD Solutions for Over-Parameterized Neural Networks.” In
Proceedings of the 37th International Conference on Machine Learning, 119:8773–84.
ML Research Press, 2020.
ieee: A. Shevchenko and M. Mondelli, “Landscape connectivity and dropout stability
of SGD solutions for over-parameterized neural networks,” in Proceedings of
the 37th International Conference on Machine Learning, 2020, vol. 119, pp.
8773–8784.
ista: Shevchenko A, Mondelli M. 2020. Landscape connectivity and dropout stability
of SGD solutions for over-parameterized neural networks. Proceedings of the 37th
International Conference on Machine Learning. vol. 119, 8773–8784.
mla: Shevchenko, Aleksandr, and Marco Mondelli. “Landscape Connectivity and Dropout
Stability of SGD Solutions for Over-Parameterized Neural Networks.” Proceedings
of the 37th International Conference on Machine Learning, vol. 119, ML Research
Press, 2020, pp. 8773–84.
short: A. Shevchenko, M. Mondelli, in:, Proceedings of the 37th International Conference
on Machine Learning, ML Research Press, 2020, pp. 8773–8784.
date_created: 2021-02-25T09:36:22Z
date_published: 2020-07-13T00:00:00Z
date_updated: 2026-10-22T22:31:18Z
day: '13'
ddc:
- '000'
department:
- _id: MaMo
- _id: DaAl
external_id:
arxiv:
- '1912.10095'
file:
- access_level: open_access
checksum: f042c8d4316bd87c6361aa76f1fbdbbe
content_type: application/pdf
creator: dernst
date_created: 2021-03-02T15:38:14Z
date_updated: 2021-03-02T15:38:14Z
file_id: '9217'
file_name: 2020_PMLR_Shevchenko.pdf
file_size: 5336380
relation: main_file
success: 1
file_date_updated: 2021-03-02T15:38:14Z
has_accepted_license: '1'
intvolume: ' 119'
language:
- iso: eng
month: '07'
oa: 1
oa_version: Published Version
page: 8773-8784
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: Proceedings of the 37th International Conference on Machine Learning
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
related_material:
record:
- id: '17465'
relation: dissertation_contains
status: public
status: public
title: Landscape connectivity and dropout stability of SGD solutions for over-parameterized
neural networks
type: conference
user_id: 8b945eb4-e2f2-11eb-945a-df72226e66a9
volume: 119
year: '2020'
...
---
_id: '6750'
abstract:
- lang: eng
text: 'Polar codes have gained extensive attention during the past few years and
recently they have been selected for the next generation of wireless communications
standards (5G). Successive-cancellation-based (SC-based) decoders, such as SC
list (SCL) and SC flip (SCF), provide a reasonable error performance for polar
codes at the cost of low decoding speed. Fast SC-based decoders, such as Fast-SSC,
Fast-SSCL, and Fast-SSCF, identify the special constituent codes in a polar code
graph off-line, produce a list of operations, store the list in memory, and feed
the list to the decoder to decode the constituent codes in order efficiently,
thus increasing the decoding speed. However, the list of operations is dependent
on the code rate and as the rate changes, a new list is produced, making fast
SC-based decoders not rate-flexible. In this paper, we propose a completely rate-flexible
fast SC-based decoder by creating the list of operations directly in hardware,
with low implementation complexity. We further propose a hardware architecture
implementing the proposed method and show that the area occupation of the rate-flexible
fast SC-based decoder in this paper is only 38% of the total area of the memory-based
base-line decoder when 5G code rates are supported. '
article_number: '8854897'
article_processing_charge: No
article_type: original
arxiv: 1
author:
- first_name: Seyyed Ali
full_name: Hashemi, Seyyed Ali
last_name: Hashemi
- first_name: Carlo
full_name: Condo, Carlo
last_name: Condo
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Warren J
full_name: Gross, Warren J
last_name: Gross
citation:
ama: Hashemi SA, Condo C, Mondelli M, Gross WJ. Rate-flexible fast polar decoders.
IEEE Transactions on Signal Processing. 2019;67(22). doi:10.1109/TSP.2019.2944738
apa: Hashemi, S. A., Condo, C., Mondelli, M., & Gross, W. J. (2019). Rate-flexible
fast polar decoders. IEEE Transactions on Signal Processing. IEEE. https://doi.org/10.1109/TSP.2019.2944738
chicago: Hashemi, Seyyed Ali, Carlo Condo, Marco Mondelli, and Warren J Gross. “Rate-Flexible
Fast Polar Decoders.” IEEE Transactions on Signal Processing. IEEE, 2019.
https://doi.org/10.1109/TSP.2019.2944738.
ieee: S. A. Hashemi, C. Condo, M. Mondelli, and W. J. Gross, “Rate-flexible fast
polar decoders,” IEEE Transactions on Signal Processing, vol. 67, no. 22.
IEEE, 2019.
ista: Hashemi SA, Condo C, Mondelli M, Gross WJ. 2019. Rate-flexible fast polar
decoders. IEEE Transactions on Signal Processing. 67(22), 8854897.
mla: Hashemi, Seyyed Ali, et al. “Rate-Flexible Fast Polar Decoders.” IEEE Transactions
on Signal Processing, vol. 67, no. 22, 8854897, IEEE, 2019, doi:10.1109/TSP.2019.2944738.
short: S.A. Hashemi, C. Condo, M. Mondelli, W.J. Gross, IEEE Transactions on Signal
Processing 67 (2019).
date_created: 2019-07-31T09:51:14Z
date_published: 2019-11-15T00:00:00Z
date_updated: 2025-01-14T14:34:26Z
day: '15'
department:
- _id: MaMo
doi: 10.1109/TSP.2019.2944738
external_id:
arxiv:
- '1903.09203'
fulldoi: https://doi.org/10.1109/TSP.2019.2944738
intvolume: ' 67'
issue: '22'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/1903.09203
month: '11'
oa: 1
oa_version: Preprint
publication: IEEE Transactions on Signal Processing
publication_identifier:
issn:
- 1053-587X
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Rate-flexible fast polar decoders
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 67
year: '2019'
...
---
_id: '7007'
abstract:
- lang: eng
text: 'We consider the primitive relay channel, where the source sends a message
to the relay and to the destination, and the relay helps the communication by
transmitting an additional message to the destination via a separate channel.
Two well-known coding techniques have been introduced for this setting: decode-and-forward
and compress-and-forward. In decode-and-forward, the relay completely decodes
the message and sends some information to the destination; in compress-and-forward,
the relay does not decode, and it sends a compressed version of the received signal
to the destination using Wyner–Ziv coding. In this paper, we present a novel coding
paradigm that provides an improved achievable rate for the primitive relay channel.
The idea is to combine compress-and-forward and decode-and-forward via a chaining
construction. We transmit over pairs of blocks: in the first block, we use compress-and-forward;
and, in the second block, we use decode-and-forward. More specifically, in the
first block, the relay does not decode, it compresses the received signal via
Wyner–Ziv, and it sends only part of the compression to the destination. In the
second block, the relay completely decodes the message, it sends some information
to the destination, and it also sends the remaining part of the compression coming
from the first block. By doing so, we are able to strictly outperform both compress-and-forward
and decode-and-forward. Note that the proposed coding scheme can be implemented
with polar codes. As such, it has the typical attractive properties of polar coding
schemes, namely, quasi-linear encoding and decoding complexity, and error probability
that decays at super-polynomial speed. As a running example, we take into account
the special case of the erasure relay channel, and we provide a comparison between
the rates achievable by our proposed scheme and the existing upper and lower bounds.'
article_number: '218'
article_type: original
arxiv: 1
author:
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: S. Hamed
full_name: Hassani, S. Hamed
last_name: Hassani
- first_name: Rüdiger
full_name: Urbanke, Rüdiger
last_name: Urbanke
citation:
ama: Mondelli M, Hassani SH, Urbanke R. A new coding paradigm for the primitive
relay channel. Algorithms. 2019;12(10). doi:10.3390/a12100218
apa: Mondelli, M., Hassani, S. H., & Urbanke, R. (2019). A new coding paradigm
for the primitive relay channel. Algorithms. MDPI. https://doi.org/10.3390/a12100218
chicago: Mondelli, Marco, S. Hamed Hassani, and Rüdiger Urbanke. “A New Coding Paradigm
for the Primitive Relay Channel.” Algorithms. MDPI, 2019. https://doi.org/10.3390/a12100218.
ieee: M. Mondelli, S. H. Hassani, and R. Urbanke, “A new coding paradigm for the
primitive relay channel,” Algorithms, vol. 12, no. 10. MDPI, 2019.
ista: Mondelli M, Hassani SH, Urbanke R. 2019. A new coding paradigm for the primitive
relay channel. Algorithms. 12(10), 218.
mla: Mondelli, Marco, et al. “A New Coding Paradigm for the Primitive Relay Channel.”
Algorithms, vol. 12, no. 10, 218, MDPI, 2019, doi:10.3390/a12100218.
short: M. Mondelli, S.H. Hassani, R. Urbanke, Algorithms 12 (2019).
corr_author: '1'
date_created: 2019-11-12T14:46:19Z
date_published: 2019-10-18T00:00:00Z
date_updated: 2024-10-09T20:59:05Z
day: '18'
ddc:
- '510'
department:
- _id: MaMo
doi: 10.3390/a12100218
external_id:
arxiv:
- '1801.03153'
file:
- access_level: open_access
checksum: 267756d8f9db572f496cd1663c89d59a
content_type: application/pdf
creator: dernst
date_created: 2019-11-12T14:48:45Z
date_updated: 2020-07-14T12:47:47Z
file_id: '7008'
file_name: 2019_Algorithms_Mondelli.pdf
file_size: 696791
relation: main_file
file_date_updated: 2020-07-14T12:47:47Z
fulldoi: https://doi.org/10.3390/a12100218
has_accepted_license: '1'
intvolume: ' 12'
issue: '10'
language:
- iso: eng
month: '10'
oa: 1
oa_version: Published Version
publication: Algorithms
publication_identifier:
issn:
- 1999-4893
publication_status: published
publisher: MDPI
quality_controlled: '1'
related_material:
record:
- id: '6675'
relation: earlier_version
status: public
scopus_import: 1
status: public
title: A new coding paradigm for the primitive relay channel
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 12
year: '2019'
...