BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Mehtaab Sawhney (MIT)
DTSTART:20210930T213000Z
DTEND:20210930T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/1
DESCRIPTION:Title: ONLINE VECTOR BALANCING\nby Mehtaab Sawhney (MIT) as part of MIT
Simple Person's Applied Mathematics Seminar\n\nLecture held in Room: 2 - 1
32 in the Simons Building.\n\nAbstract\nWe study discrepancy minimization
for vectors in $mb{R}^n$ in\nan online setting. The main result is a pair
of nearly-linear time\nonline algorithms which maintain logarithmic bounds
for the\nKoml'{o}s~conjecture. The first algorithm is based on a high-dim
ensional\ncomparison argument while the second is based on a discrete rand
om walk\non the real line with the Gaussian distribution as a stationary\n
distribution. (Joint w/ Ryan Alweiss\, Yang P. Liu\, Ashwin Sah)\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Aaron Berger (MIT)
DTSTART:20211007T213000Z
DTEND:20211007T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/2
DESCRIPTION:Title: Real Roots of Random Polynomials\nby Aaron Berger (MIT) as part o
f MIT Simple Person's Applied Mathematics Seminar\n\nLecture held in Room:
2 - 132 in the Simons Building.\n\nAbstract\nHow many real roots does a r
andom polynomial of degree n have? By the fundamental theorem of algebra\,
it can have at most n\, but in expectation the answer is often much lower
. We'll attack this problem with just a single algebraic tool (Descartes'
law of signs) and a modest helping of some probabilistic combinatorics.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Chen Lu (MIT Mathematics)
DTSTART:20211014T213000Z
DTEND:20211014T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/3
DESCRIPTION:Title: Sampling lower bounds\nby Chen Lu (MIT Mathematics) as part of MI
T Simple Person's Applied Mathematics Seminar\n\nLecture held in Room: 2 -
132 in the Simons Building.\n\nAbstract\nWe will discuss the following pr
oblem: what is the smallest number of queries we need to make to a target
distribution in order to sample from it? We will present some progress tow
ards this problem (based on joint work w/ Sinho Chewi\, Patrik Gerber\, Th
ibaut Le Gouic\, and Philippe Rigollet).\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:William Kuszmaul (MIT Mathematics)
DTSTART:20211028T213000Z
DTEND:20211028T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/4
DESCRIPTION:Title: Linear Probing Revisited: Tombstones Mark the Demise of Primary Clust
ering\nby William Kuszmaul (MIT Mathematics) as part of MIT Simple Per
son's Applied Mathematics Seminar\n\nLecture held in Room: 2 - 132 in the
Simons Building.\n\nAbstract\nThe linear-probing hash table is one of the
oldest and most widely used data structures in computer science. However\,
linear probing also famously comes with a major drawback: as soon as the
hash table reaches a high memory utilization\, elements within the hash ta
ble begin to cluster together\, causing insertions to become slow. This ph
enomenon\, now known as "primary clustering"\, was first captured by Donal
d Knuth in 1963\; at a load factor of $1 - 1/x$\, the expected time per in
sertion becomes $\\Theta(x^2)$\, rather than the more desirable $\\Theta(x
)$.\n\nWe show that there is more to the story than the classic analysis w
ould seem to suggest. It turns out that small design decisions in how dele
tions are implemented have dramatic effects on the asymptotic performance
of insertions. If these design decisions are made correctly\, then even a
hash table that is continuously at a load factor $1 - \\Theta(1/x)$ can ac
hieve average insertion time $\\tilde{O}(x)$. A key insight is that the to
mbstones left behind by deletions cause a surprisingly strong "anti-cluste
ring" effect\, and that when insertions and deletions are one-for-one\, th
e anti-clustering effects of deletions actually overpower the clustering e
ffects of insertions.\n\nBased on joint work with Michael A. Bender and Br
adley C. Kuszmaul.\nhttps://arxiv.org/abs/2107.01250\nTo appear in FOCS 20
21.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Matthew Nicoletti (MIT Mathematics)
DTSTART:20211104T213000Z
DTEND:20211104T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/5
DESCRIPTION:Title: Hydrodynamic limit for a surface growth model\nby Matthew Nicolet
ti (MIT Mathematics) as part of MIT Simple Person's Applied Mathematics Se
minar\n\nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract
\nWhether random or deterministic\, models for surface growth are studied
in many areas of pure and applied math\, and are crucial for understanding
natural phenomena. We construct and study the basic properties of a surfa
ce growth model in 2+1 dimensions.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Allen Liu (MIT EECS)
DTSTART:20211118T230000Z
DTEND:20211119T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/6
DESCRIPTION:Title: Tensor Completion Made Practical\nby Allen Liu (MIT EECS) as part
of MIT Simple Person's Applied Mathematics Seminar\n\nLecture held in Roo
m: 2 - 132 in the Simons Building.\n\nAbstract\nTensor completion is a nat
ural higher-order generalization of matrix completion where the goal is to
recover a low-rank tensor from sparse observations of its entries. Existi
ng algorithms are either heuristic without provable guarantees\, based on
solving large semidefinite programs which are impractical to run\, or make
strong assumptions such as requiring the factors to be nearly orthogonal.
In this paper we introduce a new variant of alternating minimization\, wh
ich in turn is inspired by understanding how the progress measures that gu
ide convergence of alternating minimization in the matrix setting need to
be adapted to the tensor setting. We show strong provable guarantees\, inc
luding showing that our algorithm converges linearly to the true tensors e
ven when the factors are highly correlated and can be implemented in nearl
y linear time. Moreover our algorithm is also highly practical and we show
that we can complete third order tensors with a thousand dimensions from
observing a tiny fraction of its entries. In contrast\, and somewhat surpr
isingly\, we show that the standard version of alternating minimization\,
without our new twist\, can converge at a drastically slower rate in pract
ice.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/6/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mehtaab Sawhney (MIT Mathematics)
DTSTART:20211202T230000Z
DTEND:20211203T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/7
DESCRIPTION:Title: On the Hard Core Model and Enumerating Independent Sets\nby Mehta
ab Sawhney (MIT Mathematics) as part of MIT Simple Person's Applied Mathem
atics Seminar\n\nLecture held in Room: 2 - 132 in the Simons Building.\n\n
Abstract\nSeminal results of Weitz (2005) and Sly (2010) prove that one ca
n in polynomial time approximately count independent sets in 5-regular gra
phs but cannot approximately count independent sets in 6-regular graphs (u
nless NP=RP). We discuss these results in the broader context of sampling
from the hard core model and give a high level idea of the proof of each o
f these results.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/7/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jonathan Tidor (MIT)
DTSTART:20220210T223000Z
DTEND:20220211T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/8
DESCRIPTION:Title: Communication complexity\nby Jonathan Tidor (MIT) as part of MIT
Simple Person's Applied Mathematics Seminar\n\nLecture held in Room: 2 - 1
32 in the Simons Building.\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/8/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Guanghao Ye (MIT)
DTSTART:20220217T223000Z
DTEND:20220218T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/9
DESCRIPTION:Title: Nested Dissection Meets IPMs: Planar Min-Cost Flow in Nearly-Linear T
ime\nby Guanghao Ye (MIT) as part of MIT Simple Person's Applied Mathe
matics Seminar\n\nLecture held in Room: 2 - 132 in the Simons Building.\n\
nAbstract\nWe present a nearly-linear time algorithm for finding a minimum
-cost flow in planar graphs with polynomially bounded integer costs and ca
pacities. The previous fastest algorithm for this problem is based on inte
rior-point methods (IPMs) and works for general sparse graphs in $O(n^{1.5
}\\text{poly}(\\log n))$ time [Daitch-Spielman\, STOC'08]. Our results imm
ediately extend to all families of separable graphs. This is joint work wi
th Sally Dong\, Yu Gao\, Gramoz Goranci\, Yin Tat Lee\, Richard Peng\, and
Sushant Sachdeva.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/9/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nitya Mani (MIT Mathematics)
DTSTART:20220224T223000Z
DTEND:20220225T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/10
DESCRIPTION:Title: Nim and variants\nby Nitya Mani (MIT Mathematics) as part of MIT
Simple Person's Applied Mathematics Seminar\n\nLecture held in Room: 2 -
132 in the Simons Building.\n\nAbstract\nNim is perhaps the most famous of
impartial combinatorial games\, with a simple solution initially given by
Bouton. Many generalizations of Nim and other impartial combinatorial gam
es have been closely studied since the work of Bouton. We will survey a co
llection of interesting Nim variants and some associated results\, along t
he way encountering Fibonacci numbers\, simple solutions\, and surprisingl
y complicated dynamics.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/10/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Rachel Zhang (MIT)
DTSTART:20220303T230000Z
DTEND:20220303T234500Z
DTSTAMP:20260404T110913Z
UID:SPAMS/11
DESCRIPTION:Title: Interactive Error Correcting Codes\nby Rachel Zhang (MIT) as par
t of MIT Simple Person's Applied Mathematics Seminar\n\nLecture held in Ro
om: 2 - 132 in the Simons Building.\n\nAbstract\nConsider the task of comm
unicating a message x to a receiver in an error resilient way. Classically
\, error correcting codes provide a non-interactive solution to this probl
em: the sender can simply encode x using an error correcting code\, so tha
t even if a constant fraction of the bits are adversarially corrupted\, th
e receiver can still correctly learn x. In this talk\, I will define the n
otion of an interactive error correcting code and show that over a binary
alphabet\, they can tolerate more adversarial erasures than can (non-inter
active) error correcting codes. This is joint work with Meghal Gupta and Y
ael Tauman Kalai.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/11/
END:VEVENT
BEGIN:VEVENT
SUMMARY:George Stepaniants (MIT)
DTSTART:20220310T223000Z
DTEND:20220311T000000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/12
DESCRIPTION:Title: Learning and predicting complex systems dynamics from single-variabl
e observations\nby George Stepaniants (MIT) as part of MIT Simple Pers
on's Applied Mathematics Seminar\n\nLecture held in Room: 2 - 132 in the S
imons Building.\n\nAbstract\nAdvances in model inference and data-driven s
cience have enabled the accurate discovery of governing equations from obs
ervations alone\, accelerating our understanding and control of dynamical
systems. However\, despite the ever-growing amount of experimental data co
llected\, many physical and biological systems can only be partially obser
ved. Here\, building on recent progress in the inference and integration o
f nonlinear differential equations\, we introduce an approach to learn a m
odel using observations of just a single variable within a multi-variable
dynamical system\, and use this model to accurately predict future dynamic
s. Furthermore\, we validate our approach on a variety of physical\, chemi
cal and biological systems which exhibit nonlinear dynamics such as relaxa
tion oscillations and limit cycles. This is joint work with Alasdair Haste
well\, Dominic Skinner\, Jan Totz and Jörn Dunkel.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/12/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Alex Cohen (MIT Mathematics)
DTSTART:20220331T213000Z
DTEND:20220331T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/13
DESCRIPTION:Title: A discrete 2D fractal uncertainty principle\nby Alex Cohen (MIT
Mathematics) as part of MIT Simple Person's Applied Mathematics Seminar\n\
nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\nA frac
tal uncertainty principle (FUP) states that a function `f' and its Fourier
transform cannot both be large on a fractal set. These were recently intr
oduced by Semyon Dyatlov and collaborators in order to prove new results i
n quantum chaos. So far FUPs are only understood for fractal sets in R\, a
nd fractal sets in $R^2$ remain elusive. In this talk\, we prove a sharp f
ractal uncertainty principle for Cantor sets in Z/NZ x Z/NZ\, a discrete m
odel for $R^2$.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/13/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Andrey Khesin (MIT Mathematics)
DTSTART:20220407T213000Z
DTEND:20220407T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/14
DESCRIPTION:Title: Bound Secrecy and Bound Entanglement: Where (Qu)Bits Go to Die\n
by Andrey Khesin (MIT Mathematics) as part of MIT Simple Person's Applied
Mathematics Seminar\n\nLecture held in Room: 2 - 132 in the Simons Buildin
g.\n\nAbstract\nThere is a surprising result in information theory where t
he quantum version of a conjecture is known to be true\, and the classical
one remains open. The conjecture is that there exists a joint probability
distribution for three parties\, Alice\, Bob\, and Eve\, that exhibits bo
und secrecy. Simply put\, bound secrecy is the idea that there can secret
information present in the correlations of the random variables belonging
to Alice and Bob but that is completely unknown to Eve\, and yet despite t
his\, no matter what Alice and Bob say to each other publicly\, they will
be unable to distill any bits of a secret key\, random bits completely unk
nown to Eve. This is a very surprising fact\, as it seems that there is a
secret that Alice and Bob share and yet cannot access despite their best e
fforts. The quantum version of this conjecture states that there exist joi
nt states for Alice and Bob which are entangled and therefore cannot be cr
eated without spending some amount of entanglement\, but from which no pur
e entangled states\, such as Bell pairs\, can be distilled. These joint st
ates are called bound entangled\, and not only are they known to exist\, s
ome small examples have been found.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/14/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Byron Chin (MIT Mathematics)
DTSTART:20220414T213000Z
DTEND:20220414T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/15
DESCRIPTION:Title: Reconstruction on the stochastic block model\nby Byron Chin (MIT
Mathematics) as part of MIT Simple Person's Applied Mathematics Seminar\n
\nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\nThe p
roblem of community detection is one of great relevance in machine learnin
g and data science. The goal is to group vertices that behave similarly wi
thin a graph or network. In the context of this problem\, the Stochastic B
lock Model is one of the most well-studied random graph models. In this ta
lk\, we discuss reconstruction results for this model\, highlighting a pro
vably optimal algorithm that is closely related to belief propagation.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/15/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Brice Huang (MIT Mathematics)
DTSTART:20220421T213000Z
DTEND:20220421T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/16
DESCRIPTION:Title: Tight Algorithmic Thresholds from the Branching Overlap Gap Property
\nby Brice Huang (MIT Mathematics) as part of MIT Simple Person's Appl
ied Mathematics Seminar\n\nLecture held in Room: 2 - 132 in the Simons Bui
lding.\n\nAbstract\nWe will describe recent progress on algorithmic hardne
ss results for ``spin glass-like" random optimization problems. The landsc
apes of these problems are highly nonconvex and often include many bad loc
al maxima\, impeding optimization by efficient algorithms. As a result\, t
hese problems exhibit gaps between the largest objectives that exist and t
hat can be found by efficient algorithms\; characterizing the algorithmic
limit requires developing sharp hardness results. For mean field spin glas
ses\, we describe a new hardness result that tightly characterizes the alg
orithmic limit for any algorithm with $O(1)$-Lipschitz dependence on the d
isorder coefficients. This class includes the best algorithm known for thi
s problem and natural formulations of gradient descent and approximate mes
sage passing. \\\\\n \nWe will go over the Overlap Gap Property (OGP) fram
ework of Gamarnik and Sudan\, which shows hardness in random optimization
problems against any algorithm that is suitably stable in its inputs. We i
ntroduce the \\emph{Branching OGP}\, a generalization of the OGP to arbitr
ary ultrametric constellations of solutions\, which is the key tool in the
proof of the aforementioned hardness result. By considering the Continuou
s Random Energy Model (CREM)\, we will see by analogy why the Branching OG
P yields tight algorithmic thresholds in mean field spin glasses and poten
tially in greater generality. \\\\\n \nBased on joint works with Guy Bresl
er and Mark Sellke.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/16/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Anders Aamand (MIT CSAIL)
DTSTART:20220428T220000Z
DTEND:20220428T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/17
DESCRIPTION:Title: Online Sorting and its Connection to Geometric Packing Problems\
nby Anders Aamand (MIT CSAIL) as part of MIT Simple Person's Applied Mathe
matics Seminar\n\nLecture held in Room: 2 - 132 in the Simons Building.\n\
nAbstract\nWe investigate various online packing problems in which convex
polygons arrive one by one and have to be placed irrevocably into a contai
ner before the next piece is revealed\; the pieces must not be rotated\, b
ut only translated. The aim is to minimize the used space depending on the
specific problem at hand\, e.g.\, the strip length in strip packing\, the
number of bins in bin packing\, etc. \n\n\nWe draw interesting connection
s to the following online sorting problem: We receive a stream of real num
bers $s_1\, \\cdots \,s_n\,$ each from the unit interval $[0\,1]\,$ one by
one. Each real must be placed in an array with n initially empty cells wi
thout knowing the subsequent arriving reals. The goal is to minimize the s
um of differences of consecutive reals in A. The offline optimum is to pla
ce the reals in sorted order\, so the cost is at most 1. We will see that
no online algorithm can achieve a cost lower than $n^{(1/2)}$ in the worst
case. \n\n\nWe use such lower bound to answer several fundamental questio
ns about packing. Specifically\, we prove the non-existence of competitive
algorithms for various online packing problems. A surprising corollary is
that no online algorithm can pack translates of convex polygons into a un
it square even if we are guaranteed that the total area of the pieces is a
t most 0.000001 and the diameter of each piece is at most 0.000001.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/17/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Travis Dillon (MIT Mathematics)
DTSTART:20220505T220000Z
DTEND:20220505T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/18
DESCRIPTION:Title: Fair partitioning? It's a piece of cake!\nby Travis Dillon (MIT
Mathematics) as part of MIT Simple Person's Applied Mathematics Seminar\n\
nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\nIs it
always possible to cut a cake and distribute the pieces so that everyone a
t a party gets the piece they most desire? We'll show that when it comes t
o cutting cakes\, the old adage that you can't please everybody is dead wr
ong: Everyone can have their cake and eat it\, too. Also\, we'll prove Bro
uwer's fixed point theorem without using any topology.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/18/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Shyam Narayanan (MIT EECS)
DTSTART:20221006T213000Z
DTEND:20221006T230000Z
DTSTAMP:20260404T110913Z
UID:SPAMS/19
DESCRIPTION:Title: An Introduction to Differentially Private Statistics\nby Shyam N
arayanan (MIT EECS) as part of MIT Simple Person's Applied Mathematics Sem
inar\n\nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\
nIn today's era of massive data\, various scientific and technological end
eavors have relied on machine learning or statistics models trained on use
rs (e.g.\, medical results from patient data\, better advertisement algori
thms from phone data\, etc.). Differential Privacy has recently emerged as
one of the most popular methods to protect the privacy of users. In this
talk\, I will be giving an overview of differential privacy and will focus
on how we can solve various statistical problems\, such as estimating the
mean and covariance of Multivariate Gaussian distributions\, with differe
ntial privacy using few samples. If time permits\, I will describe some mo
re recent advanced methods of solving these problems with even fewer sampl
es.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/19/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Brice Huang (MIT EECS)
DTSTART:20221013T220000Z
DTEND:20221013T224500Z
DTSTAMP:20260404T110913Z
UID:SPAMS/20
DESCRIPTION:Title: An Introduction to Random Optimization Problems\nby Brice Huang
(MIT EECS) as part of MIT Simple Person's Applied Mathematics Seminar\n\nL
ecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\nOptimiza
tion lies at the heart of many computation tasks\, such as training a neur
al network. These optimization problems arise from stochastic data and are
often high-dimensional and non-convex. In this talk\, I will overview rec
ently developed algorithms for one random optimization problem\, the mixed
p-spin glass\, and how they generalize to other problems. If time permits
I will discuss recent lower bounds suggesting that these algorithms are o
ptimal.\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/20/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lichen Zhang (MIT Mathematics)
DTSTART:20221027T220000Z
DTEND:20221027T224500Z
DTSTAMP:20260404T110913Z
UID:SPAMS/21
DESCRIPTION:Title: Sketching as a tool for fast optimization\nby Lichen Zhang (MIT
Mathematics) as part of MIT Simple Person's Applied Mathematics Seminar\n\
nLecture held in Room: 2 - 132 in the Simons Building.\n\nAbstract\n\\noin
dent Sketching is a powerful tool with many applications including regress
ion\, low-rank approximation\, and preconditioning. Given an n-by-d matrix
A with n much larger than d\, sketching describes a distribution on rando
m matrices such that an element S of this distribution is an s-by-n matrix
such that for any d-dimensional $ vector x\, || SAx ||_2^2 <= (1+eps) ||
Ax ||_2^2$ \, and $ s = poly(d\, 1/eps\, 1/delta) $ is independent of n.\n
\n\\vspace{1ex}\n\n\\noindent In this talk\, I survey another direction th
at uses sketching to speed up optimization algorithms. I’ll show how to
design a good distribution of sketching matrices so that they can be used
for \\\\\n1). Compressed gradient descent\, \n2). Linear programming and e
mpirical risk minimization. \\\\\n\n\\vspace{1ex}\n\n\\noindent We will se
e how sketching enables us to develop novel data structures for numerical
linear algebra problems\, which are the backbones of recent breakthroughs
in fast optimization algorithms. If time permits\, I’ll also touch on us
ing differential privacy (in a very surprising way) to improve the perform
ance of the sketching data structure. \\\\\n
LOCATION:https://stable.researchseminars.org/talk/SPAMS/21/
END:VEVENT
END:VCALENDAR