Search Results for author: Vipul Arora

Found 15 papers, 3 papers with code

Outlier Robust Multivariate Polynomial Regression

no code implementations14 Mar 2024 Vipul Arora, Arnab Bhattacharyya, Mathews Boban, Venkatesan Guruswami, Esty Kelman

Furthermore, we show that it is possible to have the run-time be independent of $1/\sigma$, at the cost of a higher sample complexity.

regression

Interactive singing melody extraction based on active adaptation

no code implementations12 Feb 2024 Kavya Ranjan Saxena, Vipul Arora

The proposed method is model-agnostic and hence can be applied to other non-adaptive melody extraction models to boost their performance.

Information Retrieval Melody Extraction +2

AdvNF: Reducing Mode Collapse in Conditional Normalising Flows using Adversarial Learning

no code implementations29 Jan 2024 Vikas Kanaujia, Mathias S. Scheurer, Vipul Arora

Deep generative models complement Markov-chain-Monte-Carlo methods for efficiently sampling from high-dimensional distributions.

Normalising Flows

Near-Optimal Degree Testing for Bayes Nets

no code implementations13 Apr 2023 Vipul Arora, Arnab Bhattacharyya, Clément L. Canonne, Joy Qiping Yang

This paper considers the problem of testing the maximum in-degree of the Bayes net underlying an unknown probability distribution $P$ over $\{0, 1\}^n$, given sample access to $P$.

Simultaneously Learning Robust Audio Embeddings and balanced Hash codes for Query-by-Example

no code implementations20 Nov 2022 Anup Singh, Kris Demuynck, Vipul Arora

These systems deploy indexing methods, which quantize fingerprints to hash codes in an unsupervised manner to expedite the search.

Retrieval Self-Supervised Learning

Balanced Deep CCA for Bird Vocalization Detection

no code implementations17 Nov 2022 Sumit Kumar, B. Anshuman, Linus Ruettimann, Richard H. R. Hahnloser, Vipul Arora

The key objective of this work is to learn useful embeddings associated with high performance in downstream event detection tasks when labeled data is scarce and the audio events of interest (songbird vocalizations) are sparse.

Event Detection Self-Supervised Learning

Deep domain adaptation for polyphonic melody extraction

no code implementations22 Oct 2022 Kavya Ranjan Saxena, Vipul Arora

To accomplish this task using machine learning, a large amount of labeled audio data is required to train the model that predicts the pitch contour.

Domain Adaptation Information Retrieval +4

Attention-Based Audio Embeddings for Query-by-Example

no code implementations16 Oct 2022 Anup Singh, Kris Demuynck, Vipul Arora

An ideal audio retrieval system efficiently and robustly recognizes a short query snippet from an extensive database.

Contrastive Learning Retrieval

SyncNet: Using Causal Convolutions and Correlating Objective for Time Delay Estimation in Audio Signals

1 code implementation28 Mar 2022 Akshay Raina, Vipul Arora

This paper addresses the task of performing robust and reliable time-delay estimation in audio-signals in noisy and reverberating environments.

Seq2Tok: Deep Sequence Tokenizer for Retrieval

no code implementations29 Sep 2021 Adhiraj Banerjee, Vipul Arora

Seq2Tok compresses the query and target sequences into short sequences of tokens that are faster to match.

Retrieval

Few-shot calibration of low-cost air pollution (PM2.5) sensors using meta-learning

1 code implementation2 Aug 2021 Kalpit Yadav, Vipul Arora, Sonu Kumar Jha, Mohit Kumar, Sachchida Nand Tripathi

Low-cost particulate matter sensors are transforming air quality monitoring because they have lower costs and greater mobility as compared to reference monitors.

Meta-Learning Transfer Learning

Finding Prerequisite Relations between Concepts using Textbook

no code implementations20 Nov 2020 Shivam Pal, Vipul Arora, Pawan Goyal

In the current work, we present a method of finding prerequisite relations between concepts using related textbooks.

Optimal Convergence Rate in Feed Forward Neural Networks using HJB Equation

no code implementations27 Apr 2015 Vipul Arora, Laxmidhar Behera, Ajay Pratap Yadav

It is hoped that the proposed algorithm will bring in a lot of interest in researchers working in developing fast learning algorithms and global optimization.

Cannot find the paper you are looking for? You can Submit a new open access paper.