From bspang at stanford.edu Mon Dec 2 09:38:13 2019
From: bspang at stanford.edu (Bruce Spang)
Date: Mon, 2 Dec 2019 17:38:13 +0000
Subject: [theory-seminar] This week's theory seminar
Message-ID:
Hi all,
This weeks? theory seminar will be given by Matthew S Brennan talking about "The Average-Case Complexity of Counting Cliques in Erdos-Renyi Hypergraphs.?
One special request for this week: I will be out of office on Wednesday, so need some help bringing the snacks. If you?ve been wanting to try some special snack, this is the perfect opportunity to get it! Please let me know by the end of day today if you can help :)
The talk will be on Wednesday, December 4th, from 3-4pm in Gates 463A. The abstract is below
Bruce
The Average-Case Complexity of Counting Cliques in Erdos-Renyi Hypergraphs
Matthew S Brennan
The complexity of clique problems on Erdos-Renyi random graphs has become a central topic in average-case complexity. Algorithmic phase transitions in these problems have been shown to have broad connections ranging from mixing of Markov chains to information-computation gaps in high-dimensional statistics. The focus of this talk is on relating the average-case complexity of one of these problems, where the input is chosen according to a probability distribution, to the worst-case complexity of the same problem, where the input is chosen adversarially. We consider the problem of counting k-cliques in s-uniform Erdos-Renyi hypergraphs G(n, p, s) with edge density p. In the dense regime when p is a constant, we show that the average-case complexity of counting k-cliques in G(n, p, s) essentially matches its worst-case complexity. In the sparse regime when p decays polynomially in n, we show that there are faster algorithms for counting k-cliques in G(n, p, s) than there are in the worst-case and, surprisingly, nearly matching lower bounds for G(n, p, s) can be based on worst-case complexity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From nehgupta at stanford.edu Wed Dec 4 10:37:15 2019
From: nehgupta at stanford.edu (Neha Gupta)
Date: Wed, 4 Dec 2019 18:37:15 +0000
Subject: [theory-seminar] Theory Lunch 12/05 - Arun Jambulapati
Message-ID:
Hi everyone,
This Thursday at theory lunch, Arun will tell us about "Efficient Parallel Directed Reachability in Square Root Depth" (abstract below). As usual, please join us from noon - 1pm in Gates 463A. This will be the last theory lunch of the quarter.
-------------------------------
Title: Efficient Parallel Directed Reachability in Square Root Depth
Abstract: Solving single source reachability, i.e. computing all vertices reachable from a given vertex, is one of the simplest and most well-studied graph optimization and routing problems. While obtaining work-efficient parallel algorithms for this problem in undirected graphs is trivial, obtaining comparable results for directed graphs is notoriously difficult and a key barrier towards obtaining a broader range of parallel graph algorithms. In this talk I will present recent work which shows that this problem can be solved with nearly linear work and parallel depth proportional to the square root of the number of vertices in the graph, improving upon the previous best depth bound achieved by Fineman in 2018. Further, I will discuss how these result lead to the near optimal distributed algorithms for directed reachability in the CONGEST model in certain parameter regimes.
This talk is based on joint work with Yang P. Liu and Aaron Sidford.
-------------------------------
Thanks,
Neha
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From nehgupta at stanford.edu Wed Dec 4 10:37:15 2019
From: nehgupta at stanford.edu (Neha Gupta)
Date: Wed, 4 Dec 2019 18:37:15 +0000
Subject: [theory-seminar] Theory Lunch 12/05 - Arun Jambulapati
Message-ID:
Hi everyone,
This Thursday at theory lunch, Arun will tell us about "Efficient Parallel Directed Reachability in Square Root Depth" (abstract below). As usual, please join us from noon - 1pm in Gates 463A. This will be the last theory lunch of the quarter.
-------------------------------
Title: Efficient Parallel Directed Reachability in Square Root Depth
Abstract: Solving single source reachability, i.e. computing all vertices reachable from a given vertex, is one of the simplest and most well-studied graph optimization and routing problems. While obtaining work-efficient parallel algorithms for this problem in undirected graphs is trivial, obtaining comparable results for directed graphs is notoriously difficult and a key barrier towards obtaining a broader range of parallel graph algorithms. In this talk I will present recent work which shows that this problem can be solved with nearly linear work and parallel depth proportional to the square root of the number of vertices in the graph, improving upon the previous best depth bound achieved by Fineman in 2018. Further, I will discuss how these result lead to the near optimal distributed algorithms for directed reachability in the CONGEST model in certain parameter regimes.
This talk is based on joint work with Yang P. Liu and Aaron Sidford.
-------------------------------
Thanks,
Neha
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From marykw at stanford.edu Wed Dec 4 11:07:31 2019
From: marykw at stanford.edu (Mary Wootters)
Date: Wed, 4 Dec 2019 11:07:31 -0800
Subject: [theory-seminar] Assaf Naor in probability seminar, Friday 12/6
In-Reply-To:
References:
Message-ID:
Hi all,
Assaf Naor will be giving the probability seminar on 12/6 (Friday) at 4pm
in Sloan 384H. Title and abstract attached.
Best,
Mary
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Dec06-2019.pdf
Type: application/pdf
Size: 66467 bytes
Desc: not available
URL:
From bspang at stanford.edu Wed Dec 11 10:59:43 2019
From: bspang at stanford.edu (Bruce Spang)
Date: Wed, 11 Dec 2019 18:59:43 +0000
Subject: [theory-seminar] This week's theory seminar - Chaoping Xing
Message-ID: <0059DE26-DA02-4207-81D0-F963867A8E8C@stanford.edu>
Hi all,
This week we have a special bonus theory seminar! Chaoping Xing is visiting on Friday and will talk about secure multiparty communication. The talk will be Friday, December 13th at 3pm in Gates 463a, and the title and abstract are below
Hope to see you there!
Bruce
A brief survey on secure multiparty computation
Chaoping Xing
Secure multiparty computation (MPC) was formally introduced by Andrew Yao in 1982 as secure computation between two parties. The two-party case was then generalized to multi-party by Goldreich et al. The computation is based on tools such as secret sharing of all the inputs, oblivious transfer, homomorphic encryption, and zero-knowledge proofs for ensuring that all parties behave correctly. Unlike traditional cryptographic schemes, the adversary in this model may be a participant of the computation. Therefore, a secure multi-party computation protocol must tolerate malicious behavior of participants. Over years, better multi-party schemes are proposed with regard to robustness and efficiency.
MPC has wide applications in privacy preserving computation, such as decision making like voting, statistics computations. It enables to compute aggregate functions of private inputs, while hiding the inputs themselves. For example, it can be used for calculating statistics of customer data across many banks, or of patient data across many hospitals, while hiding everything except for the final desired output of the computation. In the application of voting, the identity of each individual and his/her vote can be hidden from anyone else, yet MPC ensures that each individual is able to verify that their votes are correctly recorded and contribute towards the final decision.
In this talk, I will give a brief survey on secure multiparty computation
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From bspang at stanford.edu Fri Dec 13 12:56:14 2019
From: bspang at stanford.edu (Bruce Spang)
Date: Fri, 13 Dec 2019 20:56:14 +0000
Subject: [theory-seminar] This week's theory seminar - Chaoping Xing
In-Reply-To: <0059DE26-DA02-4207-81D0-F963867A8E8C@stanford.edu>
References: <0059DE26-DA02-4207-81D0-F963867A8E8C@stanford.edu>
Message-ID: <3410F088-EF63-47C1-8285-AB1B49F08F89@stanford.edu>
A reminder that this is today at 3pm. To celebrate vacation being right around the corner, I?ll be bringing some special snacks
See you there!
Bruce
On Dec 11, 2019, at 10:59 AM, Bruce Spang wrote:
?
Hi all,
This week we have a special bonus theory seminar! Chaoping Xing is visiting on Friday and will talk about secure multiparty communication. The talk will be Friday, December 13th at 3pm in Gates 463a, and the title and abstract are below
Hope to see you there!
Bruce
A brief survey on secure multiparty computation
Chaoping Xing
Secure multiparty computation (MPC) was formally introduced by Andrew Yao in 1982 as secure computation between two parties. The two-party case was then generalized to multi-party by Goldreich et al. The computation is based on tools such as secret sharing of all the inputs, oblivious transfer, homomorphic encryption, and zero-knowledge proofs for ensuring that all parties behave correctly. Unlike traditional cryptographic schemes, the adversary in this model may be a participant of the computation. Therefore, a secure multi-party computation protocol must tolerate malicious behavior of participants. Over years, better multi-party schemes are proposed with regard to robustness and efficiency.
MPC has wide applications in privacy preserving computation, such as decision making like voting, statistics computations. It enables to compute aggregate functions of private inputs, while hiding the inputs themselves. For example, it can be used for calculating statistics of customer data across many banks, or of patient data across many hospitals, while hiding everything except for the final desired output of the computation. In the application of voting, the identity of each individual and his/her vote can be hidden from anyone else, yet MPC ensures that each individual is able to verify that their votes are correctly recorded and contribute towards the final decision.
In this talk, I will give a brief survey on secure multiparty computation
_______________________________________________
theory-seminar mailing list
theory-seminar at lists.stanford.edu
https://mailman.stanford.edu/mailman/listinfo/theory-seminar
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From aviad at cs.stanford.edu Thu Dec 19 15:31:48 2019
From: aviad at cs.stanford.edu (Aviad Rubinstein)
Date: Thu, 19 Dec 2019 15:31:48 -0800
Subject: [theory-seminar] Fwd: USC hiring in theory
In-Reply-To:
References:
Message-ID:
---------- Forwarded message ---------
From: Shaddin Dughmi
Date: Mon, Dec 16, 2019 at 2:17 PM
Subject: USC hiring in theory
To: , Jan Vondrak , <
saberi at stanford.edu>
Amin, Aviad, Jan -
USC is hiring broadly in CS theory this year, and we anticipate making
multiple tenure-track offers if we find the right candidates. I'd
appreciate your help spreading the word by forwarding this email to the
relevant mailing lists, and letting your graduating students know.
Application Link:
https://usccareers.usc.edu/job/los-angeles/open-rank-professor-of-computer-science/1209/12581644
For more on the USC theory group, see here:
https://viterbi-web.usc.edu/~cstheory/
--
Shaddin Dughmi
Associate Professor
Computer Science Department
University of Southern California (USC)
Email: shaddin at usc.edu
Web: https://viterbi-web.usc.edu/~shaddin/
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From reingold at stanford.edu Sat Dec 28 21:46:31 2019
From: reingold at stanford.edu (Omer Reingold)
Date: Sun, 29 Dec 2019 07:46:31 +0200
Subject: [theory-seminar] resources on the ToC research methods
Message-ID:
Happy holidays everyone!
I have been updating the web page of my ToC research methods course:
https://omereingold.wordpress.com/cs-353-the-practice-of-theory-research/
At the bottom you will find links that may be useful for others too.
Suggestions for additional materials are most welcomed.
Best,
Omer
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From nehgupta at stanford.edu Tue Dec 31 10:41:26 2019
From: nehgupta at stanford.edu (Neha Gupta)
Date: Tue, 31 Dec 2019 18:41:26 +0000
Subject: [theory-seminar] Theory Lunch - Talk Sign Ups
Message-ID:
Hi everyone,
Hope you are having a good break.
The theory lunch for the winter quarter would begin on 9th January.
If you would like to give a talk, please sign up in this document -
https://docs.google.com/document/d/1S0QcDMTn-JRaP1cRFRihZyeNfHUmpYLkyORKMcGIBgY/edit .
[https://lh4.googleusercontent.com/9HdEWLwVnyEp2reMghuqC7-SBScLNXwxsZsT1nxMM7rgK_ZX6NmsmAheIm7YL94qUJDB4Q=w1200-h630-p]
Stanford TheoryCS lunch
When: Thursdays 12:00pm - 1:00pm Where: Gates 463A What: 30 min. food/social + 30 min. presentations Notes: Every Thursday noon we gather together to talk about what?s new in theoretical computer science. See our homepage for more information here http://theory.stanford.edu/main/index.shtml. Wi...
docs.google.com
Thanks,
Neha
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From nehgupta at stanford.edu Tue Dec 31 10:41:26 2019
From: nehgupta at stanford.edu (Neha Gupta)
Date: Tue, 31 Dec 2019 18:41:26 +0000
Subject: [theory-seminar] Theory Lunch - Talk Sign Ups
Message-ID:
Hi everyone,
Hope you are having a good break.
The theory lunch for the winter quarter would begin on 9th January.
If you would like to give a talk, please sign up in this document -
https://docs.google.com/document/d/1S0QcDMTn-JRaP1cRFRihZyeNfHUmpYLkyORKMcGIBgY/edit .
[https://lh4.googleusercontent.com/9HdEWLwVnyEp2reMghuqC7-SBScLNXwxsZsT1nxMM7rgK_ZX6NmsmAheIm7YL94qUJDB4Q=w1200-h630-p]
Stanford TheoryCS lunch
When: Thursdays 12:00pm - 1:00pm Where: Gates 463A What: 30 min. food/social + 30 min. presentations Notes: Every Thursday noon we gather together to talk about what?s new in theoretical computer science. See our homepage for more information here http://theory.stanford.edu/main/index.shtml. Wi...
docs.google.com
Thanks,
Neha
-------------- next part --------------
An HTML attachment was scrubbed...
URL: