arXiv subjects

Sharan Vaswani

Publications and source records attributed to Sharan Vaswani.

42 records · Page 3Linked to original sources

arXiv2018

New Insights into Bootstrapping for Bandits

We investigate the use of bootstrapping in the bandit setting. We first show that the commonly used non-parametric bootstrapping (NPB) procedure can be provably inefficient and establish a near-linear lower bound on the regret incurred by it under the bandit model with Bernoulli rewards. We show that NPB with an appropriate amount of forced exploration can result in sub-linear albeit sub-optimal regret. As an alternative to NPB, we propose a weighted bootstrapping (WB) procedure. For Bernoulli rewards, WB with multiplicative exponential weights is mathematically equivalent to Thompson sampling (TS) and results in near-optimal regret bounds. Similarly, in the bandit setting with Gaussian rewards, we show that WB with additive Gaussian weights achieves near-optimal regret. Beyond these special cases, we show that WB leads to better empirical performance than TS for several reward distributions bounded on $[0,1]$. For the contextual bandit setting, we give practical guidelines that make bootstrapping simple and efficient to implement and result in good empirical performance on real-world datasets.

cs.LG↗

arXiv2017

Horde of Bandits using Gaussian Markov Random Fields

The gang of bandits (GOB) model \cite{cesa2013gang} is a recent contextual bandits framework that shares information between a set of bandit problems, related by a known (possibly noisy) graph. This model is useful in problems like recommender systems where the large number of users makes it vital to transfer information between users. Despite its effectiveness, the existing GOB model can only be applied to small problems due to its quadratic time-dependence on the number of nodes. Existing solutions to combat the scalability issue require an often-unrealistic clustering assumption. By exploiting a connection to Gaussian Markov random fields (GMRFs), we show that the GOB model can be made to scale to much larger graphs without additional assumptions. In addition, we propose a Thompson sampling algorithm which uses the recent GMRF sampling-by-perturbation technique, allowing it to scale to even larger problems (leading to a "horde" of bandits). We give regret bounds and experimental results for GOB with Thompson sampling and epoch-greedy algorithms, indicating that these methods are as good as or significantly better than ignoring the graph or adopting a clustering-based approach. Finally, when an existing graph is not available, we propose a heuristic for learning it on the fly and show promising results.

cs.LG↗

arXiv2016

Influence Maximization with Bandits

We consider the problem of \emph{influence maximization}, the problem of maximizing the number of people that become aware of a product by finding the `best' set of `seed' users to expose the product to. Most prior work on this topic assumes that we know the probability of each user influencing each other user, or we have data that lets us estimate these influences. However, this information is typically not initially available or is difficult to obtain. To avoid this assumption, we adopt a combinatorial multi-armed bandit paradigm that estimates the influence probabilities as we sequentially try different seed sets. We establish bounds on the performance of this procedure under the existing edge-level feedback as well as a novel and more realistic node-level feedback. Beyond our theoretical results, we describe a practical implementation and experimentally demonstrate its efficiency and effectiveness on four real datasets.

cs.SI↗

arXiv2016

Adaptive Influence Maximization in Social Networks: Why Commit when You can Adapt?

Most previous work on influence maximization in social networks is limited to the non-adaptive setting in which the marketer is supposed to select all of the seed users, to give free samples or discounts to, up front. A disadvantage of this setting is that the marketer is forced to select all the seeds based solely on a diffusion model. If some of the selected seeds do not perform well, there is no opportunity to course-correct. A more practical setting is the adaptive setting in which the marketer initially selects a batch of users and observes how well seeding those users leads to a diffusion of product adoptions. Based on this market feedback, she formulates a policy for choosing the remaining seeds. In this paper, we study adaptive offline strategies for two problems: (a) MAXSPREAD -- given a budget on number of seeds and a time horizon, maximize the spread of influence and (b) MINTSS -- given a time horizon and an expected number of target users to be influenced, minimize the number of seeds that will be required. In particular, we present theoretical bounds and empirical results for an adaptive strategy and quantify its practical benefit over the non-adaptive strategy. We evaluate adaptive and non-adaptive policies on three real data sets. We conclude that while benefit of going adaptive for the MAXSPREAD problem is modest, adaptive policies lead to significant savings for the MINTSS problem.

cs.SI↗

arXiv2014

Modeling Non-Progressive Phenomena for Influence Propagation

Recent work on modeling influence propagation focus on progressive models, i.e., once a node is influenced (active) the node stays in that state and cannot become inactive. However, this assumption is unrealistic in many settings where nodes can transition between active and inactive states. For instance, a user of a social network may stop using an app and become inactive, but again activate when instigated by a friend, or when the app adds a new feature or releases a new version. In this work, we study such non-progressive phenomena and propose an efficient model of influence propagation. Specifically, we model in influence propagation as a continuous-time Markov process with 2 states: active and inactive. Such a model is both highly scalable (we evaluated on graphs with over 2 million nodes), 17-20 times faster, and more accurate for estimating the spread of influence, as compared with state-of-the-art progressive models for several applications where nodes may switch states.

cs.SI↗

arXiv2013

Fast 3D Salient Region Detection in Medical Images using GPUs

Automated detection of visually salient regions is an active area of research in computer vision. Salient regions can serve as inputs for object detectors as well as inputs for region based registration algorithms. In this paper we consider the problem of speeding up computationally intensive bottom-up salient region detection in 3D medical volumes.The method uses the Kadir Brady formulation of saliency. We show that in the vicinity of a salient region, entropy is a monotonically increasing function of the degree of overlap of a candidate window with the salient region. This allows us to initialize a sparse seed-point grid as the set of tentative salient region centers and iteratively converge to the local entropy maxima, thereby reducing the computation complexity compared to the Kadir Brady approach of performing this computation at every point in the image. We propose two different approaches for achieving this. The first approach involves evaluating entropy in the four quadrants around the seed point and iteratively moving in the direction that increases entropy. The second approach we propose makes use of mean shift tracking framework to affect entropy maximizing moves. Specifically, we propose the use of uniform pmf as the target distribution to seek high entropy regions. We demonstrate the use of our algorithm on medical volumes for left ventricle detection in PET images and tumor localization in brain MR sequences.

cs.CV↗