71 citations · 168 across the 39 of their papers we have counts for
43 papers
Gradient Coreset for Federated Learning
Durga Sivasubramanian, Lokesh Nagalapatti, Rishabh Iyer +1
Federated Learning (FL) is used to learn machine learning models with data that is partitioned across multiple clients, including resource-constrained edge devices. It is therefore…
Using Early Readouts to Mediate Featural Bias in Distillation
Rishabh Tiwari, Durga Sivasubramanian, Anmol Mekala +2
Deep networks tend to learn spurious feature-label correlations in real-world supervised learning tasks. This vulnerability is aggravated in distillation, where a student model may…
Temporally Aligning Long Audio Interviews with Questions: A Case Study in Multimodal Data Integration
Piyush Singh Pasi, Karthikeya Battepati, Preethi Jyothi +3
The problem of audio-to-text alignment has seen significant amount of research using complete supervision during training. However, this is typically not in the context of long aud…
Sāmayik: A Benchmark and Dataset for English-Sanskrit Translation
Ayush Maheshwari, Ashim Gupta, Amrith Krishna +4
We release Sāmayik, a dataset of around 53,000 parallel English-Sanskrit sentences, written in contemporary prose. Sanskrit is a classical language still in sustenance and has a ri…
INGENIOUS: Using Informative Data Subsets for Efficient Pre-Training of Language Models
H S V N S Kowndinya Renduchintala, Krishnateja Killamsetty, Sumit Bhatia +4
A salient characteristic of pre-trained language models (PTLMs) is a remarkable improvement in their generalization capability and emergence of new capabilities with increasing mod…
When Do Neural Nets Outperform Boosted Trees on Tabular Data?
Duncan McElfresh, Sujay Khandagale, Jonathan Valverde +6
Tabular data is one of the most commonly used types of data in machine learning. Despite recent advances in neural nets (NNs) for tabular data, there is still an active discussion…
A Benchmark and Dataset for Post-OCR text correction in Sanskrit
Ayush Maheshwari, Nikhil Singh, Amrith Krishna +1
Sanskrit is a classical language with about 30 million extant manuscripts fit for digitisation, available in written, printed or scannedimage forms. However, it is still considered…
Speeding up NAS with Adaptive Subset Selection
Vishak Prasad C, Colin White, Paarth Jain +2
A majority of recent developments in neural architecture search (NAS) have been aimed at decreasing the computational cost of various techniques without affecting their final perfo…
Partitioned Gradient Matching-based Data Subset Selection for Compute-Efficient Robust ASR Training
Ashish Mittal, Durga Sivasubramanian, Rishabh Iyer +2
Training state-of-the-art ASR systems such as RNN-T often has a high associated financial and environmental cost. Training with a subset of training data could mitigate this proble…
AutoML for Climate Change: A Call to Action
Renbo Tu, Nicholas Roberts, Vishak Prasad +7
The challenge that climate change poses to humanity has spurred a rapidly developing field of artificial intelligence research focused on climate change applications. The climate c…
DIAGNOSE: Avoiding Out-of-distribution Data using Submodular Information Measures
Suraj Kothawade, Akshit Srivastava, Venkat Iyer +2
Avoiding out-of-distribution (OOD) data is critical for training supervised machine learning models in the medical imaging domain. Furthermore, obtaining labeled medical data is di…
CLINICAL: Targeted Active Learning for Imbalanced Medical Image Classification
Suraj Kothawade, Atharv Savarkar, Venkat Iyer +3
Training deep learning models on medical datasets that perform well for all classes is a challenging task. It is often the case that a suboptimal performance is obtained on some cl…
DICTDIS: Dictionary Constrained Disambiguation for Improved NMT
Ayush Maheshwari, Preethi Jyothi, Ganesh Ramakrishnan
Domain-specific neural machine translation (NMT) systems (e.g., in educational applications) are socially significant with the potential to help make information accessible to a di…
Counting in the 2020s: Binned Representations and Inclusive Performance Measures for Deep Crowd Counting Approaches
Sravya Vardhani Shivapuja, Ashwin Gopinath, Ayush Gupta +2
The data distribution in popular crowd counting datasets is typically heavy tailed and discontinuous. This skew affects all stages within the pipelines of deep crowd counting appro…
AUTOMATA: Gradient Based Data Subset Selection for Compute-Efficient Hyper-parameter Tuning
Krishnateja Killamsetty, Guttu Sai Abhishek, Aakriti +4
Deep neural networks have seen great success in recent years; however, training a deep model is often challenging as its performance heavily depends on the hyper-parameters used. I…
BASIL: Balanced Active Semi-supervised Learning for Class Imbalanced Datasets
Suraj Kothawade, Pavan Kumar Reddy, Ganesh Ramakrishnan +1
Current semi-supervised learning (SSL) methods assume a balance between the number of data points available for each class in both the labeled and the unlabeled data sets. However,…
Investigating Modality Bias in Audio Visual Video Parsing
Piyush Singh Pasi, Shubham Nemani, Preethi Jyothi +1
We focus on the audio-visual video parsing (AVVP) problem that involves detecting audio and visual event labels with temporal boundaries. The task is especially challenging since i…
UDAAN: Machine Learning based Post-Editing tool for Document Translation
Ayush Maheshwari, Ajay Ravindran, Venkatapathy Subramanian +1
We introduce UDAAN, an open-source post-editing tool that can reduce manual editing efforts to quickly produce publishable-standard documents in several Indic languages. UDAAN has…
Submodlib: A Submodular Optimization Library
Vishal Kaushal, Ganesh Ramakrishnan, Rishabh Iyer
Submodular functions are a special class of set functions which naturally model the notion of representativeness, diversity, coverage etc. and have been shown to be computationally…
Adaptive Mixing of Auxiliary Losses in Supervised Learning
Durga Sivasubramanian, Ayush Maheshwari, Pradeep Shenoy +2
In several supervised learning scenarios, auxiliary losses are used in order to introduce additional information or constraints into the supervised learning objective. For instance…
Error Correction in ASR using Sequence-to-Sequence Models
Samrat Dutta, Shreyansh Jain, Ayush Maheshwari +3
Post-editing in Automatic Speech Recognition (ASR) entails automatically correcting common and systematic errors produced by the ASR system. The outputs of an ASR system are largel…
DITTO: Data-efficient and Fair Targeted Subset Selection for ASR Accent Adaptation
Suraj Kothawade, Anmol Mekala, Chandra Sekhara D +4
State-of-the-art Automatic Speech Recognition (ASR) systems are known to exhibit disparate performance on varying speech accents. To improve performance on a specific target accent…
Learning to Robustly Aggregate Labeling Functions for Semi-supervised Data Programming
Ayush Maheshwari, Krishnateja Killamsetty, Ganesh Ramakrishnan +3
A critical bottleneck in supervised machine learning is the need for large amounts of labeled data which is expensive and time consuming to obtain. However, it has been shown that…
Wisdom of (Binned) Crowds: A Bayesian Stratification Paradigm for Crowd Counting
Sravya Vardhani Shivapuja, Mansi Pradeep Khamkar, Divij Bajaj +2
Datasets for training crowd counting deep networks are typically heavy-tailed in count distribution and exhibit discontinuities across the count range. As a result, the de facto st…