School of Technology and Computer Science - STCS Seminar
http://brahma.tcs.tifr.res.in/event/stcs-seminar
enIdentifying Low-dimensional Data in High-dimensional Spaces
http://brahma.tcs.tifr.res.in/events/identifying-low-dimensional-data-high-dimensional-spaces
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Anindya De</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Northwestern University<br />
Technological Institute<br />
2145 Sheridan Road<br />
Evanston, IL 60208<br />
USA</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Tuesday, 11 December 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-12-11T14:00:00+05:30">14:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-12-11T15:00:00+05:30">15:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/piyush-srivastava">Piyush Srivastava</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Abstract:</strong> Motivated by the problem of feature selection in machine learning, the problem of testing juntas, i.e., checking if a Boolean function on the n-dimensional hypercube only depends on k<<n coordinates, has attracted a lot of attention in theoretical computer science. However, in many settings, there is no obvious choice of a basis and a more meaningful question is to ask if a function only depends a k-dimensional subspace. We show that while such "linear juntas" are not testable with any finite number of queries, assuming an upper bound of s on their surface area, such functions can tested with poly(k,s) queries, i.e., independent of the ambient dimension n. We also show a poly(s) lower bound on the query complexity of any non-adaptive tester for linear-juntas showing that the dependence on s is tight up to polynomial factors. As a consequence of our upper bound, we show that intersections of a constant number of halfspaces (as well as several related concepts) are testable with constant query complexity [joint work with Elchanan Mossel (MIT) and Joe Neeman (UT Austin)].</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Tue, 20 Nov 2018 05:30:06 +0000Supriya Pottipati4080 at http://brahma.tcs.tifr.res.inUniform Geometric and Subgeometric Ergodicity for Multiclass Many Server Queues in Heavy Traffic
http://brahma.tcs.tifr.res.in/events/uniform-geometric-and-subgeometric-ergodicity-multiclass-many-server-queues-heavy-traffic
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Aristotle Arapostathis</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>University of Texas<br />
Department of Electrical and<br />
Computer Engineering<br />
Austin, Texas, United States<br />
</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Thursday, 22 November 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-11-22T16:00:00+05:30">16:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-11-22T17:00:00+05:30">17:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/sandeep-k-juneja">Sandeep K Juneja</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Abstract:</strong> We study ergodic properties of multiclass multi-server queues, which are uniform over scheduling policies, as well as the size n of the system. The system is heavily loaded in the Halfin-Whitt regime, and the scheduling policies are work-conserving and preemptive. We provide a unified approach via ‘matching’ Foster-Lyapunov equations for both the limiting diffusion and the prelimit diffusion-scaled queueing processes simultaneously. We first study the limiting controlled diffusion, and we show that if the spare capacity (safety staffing) parameter is positive, then the diffusion is exponentially ergodic uniformly over all stationary Markov controls, and the invariant probability measures have sub-exponential tails. This result is sharp, since when there is no abandonment and the spare capacity parameter is negative, then the controlled diffusion is transient under any Markov control. In addition, we show that if all the abandonment rates are positive, the invariant probability measures have sub-Gaussian tails, regardless whether the spare capacity parameter is positive or negative.</p>
<p>Using the above results, we proceed to establish the corresponding ergodic properties for the diffusion-scaled queueing processes. In addition to providing a simpler proof of the results in Gamarnik and Stolyar [Queueing Syst. (2012) 71:25--51], we extend these results to GI/M/n+M queues with renewal arrival processes, albeit under the assumption that the mean residual life functions are bounded. For the Markovian model with Poisson arrivals, we obtain stronger results and show that the convergence to the stationary distribution is at a geometric rate uniformly over all work-conserving stationary Markov scheduling policies.</p>
<p>We then turn to the case when arrivals are heavy-tailed, or the system suffers from asymptotically negligible service interruptions. In these models, the Itô equations are driven by either (1) a Brownian motion and a pure-jump Levy process, or (2) an anisotropic Levy process with independent one-dimensional symmetric alpha-stable components, or (3) an anisotropic Lévy process and a pure-jump Lévy process. We identify conditions on the parameters in the drift, the Lévy measure and/or covariance function which result in subexponential and/or exponential ergodicity. We show that these assumptions are sharp. In addition, we show that for the queueing models described above with no abandonment, the rate of convergence is polynomial, and we provide a sharp quantitative characterization of this rate via matching upper and lower bounds (joint work with Hassan Hmedi, Guodong Pang and Nikola Sandric).</p>
<p> <strong>Bio: </strong>Ari Arapostathis is a professor at the Department of Electrical and Computer Engineering at the University of Texas at Austin. His research interests include analysis and estimation techniques for stochastic systems, stability properties of large-scale interconnected power systems, and stochastic and adaptive control theory. His main technical contributions are in the areas of adaptive control and estimation of stochastic systems with partial observations, controlled diffusions, adaptive control of nonlinear systems, geometric nonlinear theory, and stability of large scale interconnected power systems. His research is currently funded by the National Science Foundation (Division of Mathematical Sciences), Army Research Office (Applied Probability), and the Office of Naval Research. He is a Fellow of IEEE.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Mon, 19 Nov 2018 11:12:43 +0000Supriya Pottipati4079 at http://brahma.tcs.tifr.res.inOn the Structure and Lower Bounds for Multilinear Branching Programs
http://brahma.tcs.tifr.res.in/events/structure-and-lower-bounds-multilinear-branching-programs
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Ramya C.</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Ph.D. student<br />
Department of Computer<br />
Science and Engineering<br />
IIT Madras</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Wednesday, 21 November 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-11-21T16:00:00+05:30">16:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-11-21T17:00:00+05:30">17:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/ramprasad-saptharishi">Ramprasad Saptharishi</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Abstract: </strong>Polynomials are the most fundamental mathematical objects in algebra and it is compelling to understand the complexity of computing polynomials. That is, given a polynomial we want to understand the number of arithmetic operations needed to compute it. Leslie G. Valiant introduced the notion of arithmetic circuits as a model for computing polynomials. We will primarily be interested in size of an arithmetic circuit computing f which is the number of arithmetic operations needed to compute the polynomial f. Further, Valiant conjectured that the permanent of an nxn matrix viewed as a polynomial cannot be computed by arithmetic circuits of size polynomial in n. Subsequent to Valiant's conjecture, proving size lower bounds for circuits computing permanent have been of much interest. While the answer to Valiant's conjecture seems to be at a distance, the focus has been on restricted classes of circuits.</p>
<p>In this talk, we will focus on Algebraic Branching Programs, yet another model for computing polynomials. Interesting polynomial families such as determinant, permanent etc. being multilinear, it is natural to consider syntactic multilinear Algebraic Branching Programs(smABPs) where every variable appears as an edge label at most once along any path in the branching program. The best known size lower bound for smABPs is barely quadratic in the number of variables. Proving super-polynomial size lower bounds for smABPs computing an explicit multilinear polynomial is a challenging problem in Algebraic Complexity Theory.</p>
<p>In this talk, we aim to understand the structure of smABPs and outline possible approaches to prove super-polynomial lower bounds for smABPs. We obtain a new decomposition theorem for smABPs: We show that any n-variate polynomial that can be computed by an smABP of size S, can be written as a sum of O(S) many multilinear polynomials where each summand is a product of two polynomials in at most 2n/3 variables, computable by smABPs. As an immediate corollary to our decomposition theorem for smABPs, we obtain a low bottom fan-in version of the depth reduction by Tavenas[MFCS, 2013] for the case of smABPs. This also leaves us with certain structural observations on smABPs which may be exploited to obtain super-polynomial lower bounds for smABPs. This is joint work with B.V.Raghavendra Rao, IIT Madras.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Mon, 19 Nov 2018 05:54:42 +0000Supriya Pottipati4076 at http://brahma.tcs.tifr.res.inData Assimilation for Deterministic High Dimensional Dynamical Systems
http://brahma.tcs.tifr.res.in/events/data-assimilation-deterministic-high-dimensional-dynamical-systems
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Amit Apte</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>International Centre for Theoretical Sciences<br />
Tata Institute of Fundamental Research<br />
Bengaluru North 560089 </p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Tuesday, 20 November 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-11-20T14:00:00+05:30">14:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-11-20T15:00:00+05:30">15:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p>Data assimilation refers to the problem of estimation of state of a high dimensional chaotic system given noisy, partial observations of the system. A variety of techniques, mainly falling in one of two broad classes of either variational or Bayesian, have been developed in the context of earth sciences, mainly for weather prediction purposes. This talk will focus on the Bayesian viewpoint (nonlinear filtering for deterministic dynamics), illustrating how the characteristics of the dynamics of the system play a crucial role in determining the properties of filtering distribution. One example of this relation is our recent work (doi:10.1137/15M1025839, doi:10.1137/16M1068712) related to the convergence of the Kalman filter covariance matrix onto the unstable-neutral subspace for a linear, deterministic dynamical system with linear observation operator, which I will discuss briefly. The second example will focus on Lagrangian data assimilation (LaDA) which refers to the use of observations provided by (pseudo-)Lagrangian instruments such as drifters, floats, and gliders, which are important sources of surface and subsurface data for the oceans. I will describe our recent proposal (doi:10.1175/MWR-D-14-00051.1, doi.org/10.1007/978-3-319-25138-7_24) for a hybrid particle-Kalman filter method for LaDA, which combines the strengths of both these filters and the specific dynamical structure of the Lagrangian dynamics.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Mon, 05 Nov 2018 09:21:02 +0000Supriya Pottipati4073 at http://brahma.tcs.tifr.res.inFairness & Diversity in Online Social Systems
http://brahma.tcs.tifr.res.in/events/fairness-diversity-online-social-systems
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Elisa Celis</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Ecole Polytechnique Federale de Lausanne (EPFL)<br />
School of Computer and Communication Sciences<br />
Switzerland<br />
</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Friday, 26 October 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-10-26T10:00:00+05:30">10:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-10-26T11:00:00+05:30">11:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/piyush-srivastava">Piyush Srivastava</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Abstract:</strong> Social systems are now fueled by algorithms that facilitate and control connections and information. Simultaneously, computational systems are now fueled by people -- their interactions, data, and behavior. Consequently, there is a pressing need to design new algorithms that are socially responsible in how they learn, and socially optimal in the manner in which they use information. Recently, we have made initial progress in addressing such problems at this interface of social and computational systems. In this talk, we will first understand the emergence of bias in data and algorithmic decision making and present first steps towards developing a systematic framework to control biases in classical problems such as data summarization and personalization. This work leads to new algorithms that have the ability to alleviate bias and increase diversity while often simultaneously maintaining their theoretical or empirical performance with respect to the original metrics.</p>
<p><strong>Bio:</strong> Elisa Celis is a Senior Research Scientist at the School of Computer and Communication Sciences at EPFL. Starting January 2019, she will be an Assistant Professor in the Department of Statistics and Data Science at Yale. Previously, she worked as a Research Scientist at Xerox Research where she was the worldwide head of the Crowdsourcing and Human Computation research thrust. She received a B. Sci. degree in Computer Science and Mathematics from Harvey Mudd College and a Ph. D. in Computer Science from the University of Washington. Her research focuses on studying social and economic questions that arise in the context of the Internet and her work spans multiple areas including fairness in AI/ML, social computing, online learning, network science, and mechanism design. She is the recipient of the Yahoo! Key Challenges Award and the China Theory Week Prize.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Wed, 24 Oct 2018 08:17:53 +0000Supriya Pottipati4070 at http://brahma.tcs.tifr.res.inAlgorithms from Physics
http://brahma.tcs.tifr.res.in/events/algorithms-physics
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Nisheeth Vishnoi</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>EPFL IC IIF THL3<br />
INJ 135 - Station 14<br />
CH-1015 Lausanne<br />
Switzerland</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Friday, 26 October 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-10-26T11:15:00+05:30">11:15</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-10-26T12:15:00+05:30">12:15</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/piyush-srivastava">Piyush Srivastava</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Abstract:</strong> In understanding physical systems over hundreds of years, physicists have developed a wealth of dynamics and viewpoints. Some of these methods, when abstracted appropriately, could lead to new algorithmic techniques with applications to machine learning and theoretical computer science. I will present a couple of recent examples from my own research on such interactions between Physics and Algorithms -- a Hamiltonian Dynamics inspired algorithm for sampling from continuous distributions and a Boltzmann's equation based algorithm for estimating the partition function for discrete distributions.</p>
<p><strong>Bio:</strong> Nisheeth Vishnoi is currently a professor in the School of Computer and Communication Sciences at École Polytechnique Fédérale de Lausanne where he heads the theory of computation lab. Starting January 2019, he will a professor of Computer Science at Yale. He is also an associate of the International Center for Theoretical Sciences, Bangalore, an adjunct faculty member of IIT Delhi and IIT Kanpur, and a co-founder of the Computation, Nature, and Society ThinkTank in Lausanne. His research focuses on foundational problems in algorithms, optimization, and statistics, and how tools from these areas can be used to address computational questions in society and other sciences. Topics from these areas that he is currently interested in include algorithmic bias and the emergence of intelligence. He is the recipient of the Best Paper Award at FOCS 2005, the IBM Research Pat Goldberg Memorial Award for 2006, the Indian National Science Academy Young Scientist Award for 2011 and the IIT Bombay Young Alumni Achievers Award for 2016.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Wed, 24 Oct 2018 06:44:09 +0000Supriya Pottipati4067 at http://brahma.tcs.tifr.res.inA Fresh Look at an Old Problem: Network Utility Maximization—Convergence, Delay, and Complexity
http://brahma.tcs.tifr.res.in/events/fresh-look-old-problem-network-utility-maximization%E2%80%94convergence-delay-and-complexity
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Ness Shroff</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>The Ohio State University<br />
Electrical and Computer Engineering<br />
Columbus, OH 43210</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Monday, 29 October 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-10-29T11:30:00+05:30">11:30</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-10-29T12:30:00+05:30">12:30</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/sandeep-k-juneja">Sandeep K Juneja</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Absract:</strong> Network Utility Maximization has been studied for resource allocation problems in communication networks for nearly two decades. Nonetheless, a major challenge that continues to remain open is how to develop a distributed congestion control and routing algorithm that can simultaneously provide utility optimality, fast convergence speed, and low delay. To address this challenge we take a fresh perspective on this old problem and develop a new algorithm that offers the fastest known convergence speed, vanishing utility optimality gap with finite queue length, and low routing complexity.</p>
<p> Our key contributions in this work are: i) the design of a new joint congestion control and routing algorithm based on a type of inexact Uzawa method in the Alternating Directional Method of Multiplier; ii) a new theoretical path to prove global and linear convergence rate without requiring the full rank assumption of the constraint matrix; and iii) a clear path for implementing the proposed method in a fully distributed fashion.</p>
<p> <strong>Bio:</strong> Ness B. Shroff received his Ph.D. degree from Columbia University, NY in 1994 and joined Purdue university immediately thereafter as an Assistant Professor. At Purdue, he became Professor of the school of Electrical and Computer Engineering and director of a university wide center on wireless systems and applications (CWSA) in 2004. In July 2007, he joined the ECE and CSE departments at The Ohio State University, where he holds the Ohio Eminent Scholar Chaired Professorship of Networking and Communications. >From 2009-2012, he also served as a Guest Chaired professor of Wireless Communications at Tsinghua University, Beijing, China, and currently holds an honorary Guest professor at Shanghai Jiaotong University in China and visiting position at the Indian Institute of Technology, Bombay.</p>
<p> Dr. Shroff's research interests span the areas of communication, networking, storage, cloud, recommender, social, and cyberphysical systems. He is especially interested in fundamental problems in learning, design, control, performance, pricing, and security of these complex systems. He currently serves as chair of the ACM Mobihoc steering committee, editor-at-large in the IEEE/ACM Trans. on Networking, and as senior editor of the IEEE Transactions on Control of Networked Systems. He also serves on the editorial boards of the IEEE Network Magazine, and the Network Science journal. He has served on the technical and executive committees of several major conferences and workshops. For example, he was the technical program co-chair of IEEE INFOCOM'03, the premier conference in communication networking, the technical program co-chair of ACM Mobihoc 2008, the General co-chair of WICON'08, and the conference chair of IEEE CCW'99. He has served as a keynote speaker and panelist on several major conferences in these fields. Dr. Shroff was also a co-organizer of the NSF workshop on Fundamental Research in Networking in 2003, and the NSF workshop on the Future of Wireless Networks in 2009.</p>
<p> Dr. Shroff is a Fellow of the IEEE, and a National Science Foundation CAREER awardee. His papers have received numerous awards at top-tier venues. For example, he received the best paper award at IEEE INFOCOM 2006, IEEE INFOCOM 2008, and IEEE INFOCOM 2016, the best paper of the year in the journal of Communication and Networking (2005) and in Computer Networks (2003). He also also received runner-up awards at IEEE INFOCOM 2005 and IEEE INFOCOM 2013. In addition, his papers have received the best student paper award (from all papers whose first author is a student) at IEEE WIOPT 2013, IEEE WiOPT 2012, and IEEE IWQoS 2006. Dr. Shroff is on the list of highly cited researchers from Thomson Reuters ISI (previously ISI web of Science) in 2014 and 2015, and in Thomson Reuters Book on The World's Most Influential Scientific Minds in 2014. He received the IEEE INFOCOM achievement award for seminal contributions to scheduling and resource allocation in wireless networks, in 2014.</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Tue, 23 Oct 2018 11:24:48 +0000Supriya Pottipati4066 at http://brahma.tcs.tifr.res.inBetti Numbers of Gaussian Excursions in the Sparse Regime
http://brahma.tcs.tifr.res.in/events/betti-numbers-gaussian-excursions-sparse-regime
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Gugan Thoppe</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Duke University<br />
Durham, USA</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Thursday, 15 November 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-11-15T16:00:00+05:30">16:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-11-15T17:00:00+05:30">17:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/sandeep-k-juneja">Sandeep K Juneja</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p>Abstract: A function's excursion set is the sub-domain where its value exceeds some threshold. Some key examples illustrating the central role that excursion sets play in different application areas are as follows. In medical imaging, in order to understand the brain parts involved in a particular task, analysts frequently look at the high blood flow level regions in the brain when the said task is being performed. In control theory, it is known that the viability and invariance properties of control systems can be expressed as super-level sets of suitable value functions. In robotics, in order to plan its motion, a sensor robot may want to identify the sub-terrain where the accessibility probability is above some threshold. Often, functions whose excursions are of interest are either random themselves (for e.g., due to noise) or, while being deterministic, are too complicated and hence can be treated as being a sample of a random field. In this sense, studying the topology of random field excursions is vital. This work is the first detailed study of their Betti numbers (number of holes) in the so-called `sparse' regime. Specifically, we consider a piecewise constant Gaussian field whose covariance function is positive and satisfies some local, boundedness, and decay rate conditions. We model its excursion set via a Cech complex. For Betti numbers of this complex, we then prove various limit theorems as the window size and the excursion level together grow to infinity. Our results include asymptotic mean and variance estimates, a vanishing to non-vanishing phase transition with a precise estimate of the transition threshold, and a weak law in the non-vanishing regime. We further have a Poisson approximation and a central limit theorem close to the transition threshold. Our proofs combine extreme value theory and combinatorial topology tools (joint work with Sunder Ram Krishnan).</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Mon, 22 Oct 2018 11:44:55 +0000Supriya Pottipati4065 at http://brahma.tcs.tifr.res.inLearning and Testing Causal Models with Interventions
http://brahma.tcs.tifr.res.in/events/learning-and-testing-causal-models-interventions
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Saravanan Kandasamy</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Indian Institute of Science<br />
Department of Computer<br />
Science & Automation<br />
Bangalore 560012</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Tuesday, 16 October 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-10-16T14:00:00+05:30">14:00</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-10-16T15:00:00+05:30">15:00</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/piyush-srivastava">Piyush Srivastava</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p>Abstract: We consider testing and learning problems on causal Bayesian networks as defined by Pearl. Given a causal Bayesian network M on a graph with n discrete variables and bounded in-degree and bounded ``confounded components'', we show that O(log n) interventions on an unknown causal Bayesian network X on the same graph, and O(n/epsilon^2) samples per intervention, suffice to efficiently distinguish whether X=M or whether there exists some intervention under which X and M are farther than epsilon in total variation distance. We also obtain sample/time/intervention efficient algorithms for: (i) testing the identity of two unknown causal Bayesian networks on the same graph; and (ii) learning a causal Bayesian network on a given graph. Although our algorithms are non-adaptive, we show that adaptivity does not help in general: Omega(log n) interventions are necessary for testing the identity of two unknown causal Bayesian networks on the same graph, even adaptively. Our algorithms are enabled by a new subadditivity inequality for the squared Hellinger distance between two causal Bayesian networks (joint work with: Jayadev Acharya, Arnab Bhattacharyya and Constantinos Daskalakis).</p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Tue, 09 Oct 2018 04:51:29 +0000Supriya Pottipati4062 at http://brahma.tcs.tifr.res.inConditional Independence : Novel Consistent Estimators & Hypothesis Testing Paradigm
http://brahma.tcs.tifr.res.in/events/conditional-independence-novel-consistent-estimators-hypothesis-testing-paradigm
<div class="field field-name-field-event-type field-type-taxonomy-term-reference field-label-hidden view-mode-rss clearfix">
<ul class="field-items">
<li class="field-item even">
<a href="/event/stcs-seminar" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">STCS Seminar</a> </li>
</ul>
</div>
<section class="field field-name-field-other field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Speaker: </h2><div class="field-items"><div class="field-item even">Himanshu Asnani</div></div></section><section class="field field-name-field-affiliation field-type-text-long field-label-inline clearfix view-mode-rss"><h2 class="field-label">Affiliation: </h2><div class="field-items"><div class="field-item even"><p>Department of Electrical Engineering<br />
University of Washington<br />
Seattle, Washington.</p>
</div></div></section><section class="field field-name-field-time field-type-datetime field-label-inline clearfix view-mode-rss"><h2 class="field-label">Time: </h2><div class="field-items"><div class="field-item even"><span class="date-display-single">Monday, 15 October 2018, <span class="date-display-range"><span class="date-display-start" property="dc:date" datatype="xsd:dateTime" content="2018-10-15T14:30:00+05:30">14:30</span> to <span class="date-display-end" property="dc:date" datatype="xsd:dateTime" content="2018-10-15T15:30:00+05:30">15:30</span></span></span></div></div></section><section class="field field-name-field-venue field-type-taxonomy-term-reference field-label-inline clearfix view-mode-rss clearfix">
<h2 class="field-label">Venue: </h2>
<ul class="field-items">
<li class="field-item even">
<a href="/venue/201-stcs-seminar-room" typeof="skos:Concept" property="rdfs:label skos:prefLabel" datatype="">A-201 (STCS Seminar Room)</a> </li>
</ul>
</section>
<section class="field field-name-field-organisers field-type-entityreference field-label-inline clearfix view-mode-rss"><h2 class="field-label">Organisers: </h2><div class="field-items"><div class="field-item even"><a href="/people/vinod-m-prabhakaran">Vinod M. Prabhakaran</a></div></div></section><div class="field field-name-body field-type-text-with-summary field-label-hidden view-mode-rss"><div class="field-items"><div class="field-item even" property="content:encoded"><p><span style="font-size:12px;"><strong>Abstract:</strong> The problem of ascertaining conditional independence or dependence is central to causal discovery and statistical inference in several dynamical systems, such as gene regulatory networks, finance networks, edge testing in Bayesian Networks etc. In this talk, we navigate this problem via two statistical approaches.</span></p>
<p><span style="font-size:12px;">In the first approach, for the low dimensional regime, we develop consistent sample estimators based on nearest neighbor methods for conditional mutual information (CMI) in general probability spaces, that is, even when the variables are mixtures of continuous and discrete components or have only low-dimensional manifolds. In general, we define a general graph divergence measure (GDM), as a measure of incompatibility between the observed distribution and a given graphical model structure. This generalizes to estimating several multivariate information measures, a special case of which is conditional mutual information. We construct a novel estimator via a coupling trick that directly estimates these multivariate information measures using the Radon-Nikodym derivative.</span></p>
<p> For the second approach, in the case of high dimensional regime, we study the conditional independence hypothesis testing problem, where we develop a new <strong>"mimic and classify"</strong> paradigm which is realized in two-steps: (a) <strong>mimic</strong> the conditionally independent (CI) distribution close enough to recover the support, and (b) <strong>classify</strong> to distinguish the joint and the CI distribution. Thus, as long as we have a good generative model and a good classifier, we potentially have a sound CI Tester. With this modular paradigm which has provable p-value guarantees, CI Testing also becomes amiable to be handled by state-of-the-art, both generative and classification methods from the modern advances in Deep Learning.</p>
<p><span style="font-size:12px;">Both the above approaches are benchmarked on synthetic and real datasets. Finally, with lessons drawn from these two approaches, we present future research program, broadly, at the intersection of Information Theory and Statistical Learning (including modern Deep Learning methods) and their applications.</span></p>
<p><span style="font-size:12px;"><strong>Bio: </strong>Dr. Himanshu Asnani is currently a Research Associate in the Electrical Engineering Department at University of Washington, Seattle and Visiting Assistant Professor in the Electrical Engineering Department at IIT Bombay. His research interests include information and coding theory, statistical learning and inference and machine learning. He has been named Amazon Catalyst Fellow for the year 2018. Dr. Asnani is the recipient of 2014 Marconi Society Paul Baran Young Scholar Award. He received his Ph.D. in Electrical Engineering Department in 2014 from Stanford University, working under Professor Tsachy Weissman, where he was a Stanford Graduate Fellow. Following his graduate studies, he worked in Ericsson Silicon Valley as a System Architect for couple of years, focusing on designing next generation networks with emphasis on network redundancy elimination and load balancing. Driven by a deep desire to innovate and contribute in the education space, with the aid of technology, Dr. Himanshu Asnani quit his corporate sojourn and got involved for a while in his education startups (where he currently holds Founding Advisor role) to bring the promise of quality education in vernacular languages in underdeveloped and developing countries - places which do not have access to English, Internet and Electricity. In the past, he has also held visiting faculty appointments in the Electrical Engineering Department, Stanford University. He was the recipient of Best Paper Award at MobiHoc 2009 and was also the finalist for Student Paper Award in ISIT 2011, Saint Petersburg, Russia. Prior to that, he received his B.Tech. from IIT Bombay in 2009 and M.S. from Stanford University in 2011, both in Electrical Engineering.</span></p>
</div></div></div><section class="field field-name-field-diplay-profile field-type-list-boolean field-label-above view-mode-rss"><h2 class="field-label">Faculty Candidate: </h2><div class="field-items"><div class="field-item even"></div></div></section>Mon, 17 Sep 2018 04:43:02 +0000Supriya Pottipati4059 at http://brahma.tcs.tifr.res.in