Ragesh Jaiswal

	I am an associate professor in the Department of Computer Science and Engineering at IIT Delhi. I am interested in Algorithms, Complexity, Theoretical Cryptography, and Machine Learning.		Office Address: Room 403, SIT Building, IIT Delhi, Hauz Khas, New Delhi India - 110 016
Before joining IIT Delhi, I worked as a Postdoctoral Research Scientist at Columbia University, New York. Before this, I did my PhD under Prof. Russell Impagliazzo from the University of California San Diego. I did my BTech from IIT Kanpur.			Email:

Teaching

• COL351: Analysis and Design of Algorithms (Semester I, 2020-21).
• COL863: Quantum Computing (Semester II, 2018-19).
• COL202: Discrete Mathematical Structures (Semester I, 2018-19).
• CSL202: Discrete Mathematical Structures (Semester II, 2017-18).
• COL866: Foundations of Data Science (Semester I, 2017-18).
• COL106: Data Structures and Algorithms (Semester II, 2016-17).
• COL351: Analysis and Design of Algorithms (Semester I, 2016-17).
• Courses taught at University of California San Diego (Fall-2015, Winter-2016).
• COL870: Special Topics in Machine Learning (Clustering Algorithms)(Semester II, 2014-15).
• CSL356: Analysis and Design of Algorithms(Semester I, 2014-15).
• CSL853: Complexity Theory(Semester II, 2013-14).
• CSL105: Discrete Mathematical Structures(Semester I, 2013-14).
• CSL759: Cryptography and Computer Security(Semester II, 2012-13).
• CSL356: Analysis and Design of Algorithms(Semester I, 2012-13).
• CSL857: Randomized Algorithms(Semester II, 2011-12).
• CSL356: Analysis and Design of Algorithms(Semester I, 2011-12).
• CSL866: Cryptography(Semester II, 2010-11).
• CSL201: Data Structures(Semester I, 2010-11).

• Anup Bhattacharya (2012-2018)
• Dishant Goyal (2017-present, co-advised by Prof. Amit Kumar)

• Anup Bhattacharya, Dishant Goyal, and Ragesh Jaiswal,
  Hardness of Approximation of Euclidean k-Median. [arXiv]
  Short Description: We give hardness of approximation result for the Euclidean k-median problem assuming the Unique Games Conjecture (UGC). This solves an open question posed in this work.
  
  
• Dishant Goyal, Ragesh Jaiswal, and Amit Kumar,
  FPT Approximation for Constrained Metric k-Median/Means. IPEC'20. [arXiv][IPEC]
  Short Description: We give (3+ε)-approximation and (9+ε)-approximation for various constrained versions of k-median and k-means problem respectively in FPT time.
  
  
• Anup Bhattacharya, Dishant Goyal, Ragesh Jaiswal, Amit Kumar,
  On Sampling Based Algorithms for k-Means. FSTTCS'20.[FSTTCS]
  Short Description: This is a combined presentation of the works below.
  
  • Anup Bhattacharya, Dishant Goyal, Ragesh Jaiswal, and Amit Kumar,
    Streaming PTAS for Binary ℓ₀-Low Rank Approximation. [arXiv]
    Short Description: We give constant-pass, polylog-space streaming PTAS for the constrained binary k-means problem and the generalised binary ℓ₀-rank-r approximation problem.
    
  • Dishant Goyal, Ragesh Jaiswal, and Amit Kumar,
    Streaming PTAS for Constrained k-Means. [arXiv]
    Short Description: We generalise our previous results and use it to obtain constant-pass, logspace streaming PTASs for a variety of constrained k-means problems as well as faster PTAS for k-means under stability.
    
    
• Ragesh Jaiswal and Amit Kumar,
  Multiplicative Rank-1 Approximation using Length-Squared Sampling. SOSA@SODA'20. [arXiv][Talk]
  Short Description: We show that the span of Ω(1/ε⁴) rows of any matrix sampled according to the length-squared distribution, contains a rank-1 matrix B that gives a (1+ε) multiplicative approximation under the Frobenius norm. Previous known results gave additive approximation.
  
  
• Ragesh Jaiswal, A note on the relation between XOR and Selective XOR Lemmas.
  [ Information Processing Letters] [arXiv]
  Short Description: We give a simple reduction from the Selective XOR lemma to the standard XOR lemma.
  
  
• Nir Ailon, Anup Bhattacharya, Ragesh Jaiswal, Amit Kumar,
  Approximate Clustering with Same-Cluster Queries. ITCS 2018. [arXiv][Talk]
  Short Description: In this work we explore the power and limitation of same-cluster queries in the context of approximate clustering.
  
  
• Ragesh Jaiswal, Sandeep Sen,
  Approximate Clustering [PDF]
  Book chapter to appear in Approximation Algorithms and Metaheuristics - Volume II
  Short Description: Contains a survey on sampling based clustering algorithms.
  
  
• Russell Impagliazzo, Ragesh Jaiswal, Valentine Kabanets, Bruce M. Kapron, Valerie King, and Stefano Tessaro,
  Simultaneous Secrecy and Reliability Amplification for a General Channel Model,
  TCC-B, 2016.[PDF]
  Short Description: We present a general notion of channel for cryptographic purposes, which can model either a (classical) physical channel or the consequences of a cryptographic protocol, or any hybrid. We consider simultaneous secrecy and reliability amplification for such channels. We show that simultaneous secrecy and reliability amplification is not possible for the most general model of channel, but, at least for some values of the parameters, it is possible for a restricted class of channels that still includes both standard information-theoretic channels and keyless cryptographic protocols.
  
  
• Anup Bhattacharya, Ragesh Jaiswal, and Amit Kumar,
  Faster Algorithms for the Constrained k-means Problem,
  STACS 2016. Theory of Computing Systems [STACS special issue].
  [arXiv] [Slides]
  Short description: The classical center based clustering problems such as k-means/median/center assume that the optimal clusters satisfy the locality property that the points in the same cluster are close to each other. A number of clustering problems arise in machine learning where the optimal clusters do not follow such a locality property. For instance, consider the r-gather clustering problem where there is an additional constraint that each of the clusters should have at least r points or the capacitated clustering problem where there is an upper bound on the cluster sizes. In this work, we consider an extreme generalisation of such problems by assuming that the target clustering is an arbitrary partition of the dataset. We show some interesting results in this scenario.
  
  
• Anup Bhattacharya, Davis Issac, Ragesh Jaiswal, and Amit Kumar,
  Sampling in Space Restricted Settings,
  COCOON 2015. Algorithmica [COCOON special issue].
  [arXiv]
  Short description: Consider the problem of sampling a uniformly random element from a data stream. This is a standard interview question and the classic solution is to store the first item and when item i (i>1) arrives, then with probability 1/i replace the currently stored element with item i and with the remaining probability, do nothing. It is simple to argue that after seeing n items, the probability that the stored element is item i is exactly 1/n for all i<=n. The space required for this algorithm is O(log n) which is optimal. However, the randomness requirement of this algorithm is O(n log n) which is large. Randomness as a resource is relevant in computation since generating (pseudo)randomness costs energy. In the first part of this work, we show that the the above goal can be achieved with O(log n) space and O(log n) random bits.