Author name cluster

Mayank Agarwal

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

EAAI Journal 2025 Journal Article

Standalone and Hybrid machine learning approaches to predict sediment load in an alluvial channel

Sanjit Kumar
Vishal Deshpande
Mayank Agarwal

A significant amount of sediment transported in an alluvial river can alter the morphology and shape of the river. Accurate prediction of sediment load is essential in studying the change in geomorphology and dynamics of rivers and also to evaluate its impact on aquatic ecosystems, infrastructure, and human activities dependent on water resources. The present study demonstrates framework for predicting sediment load in alluvial channels using both standalone and hybrid machine learning (ML) models. Multiple datasets collected from various river surveys and flume studies were used to evaluate the significance of key variables such as friction slope (Sf), channel discharge (Q), and bed shear stress (τ b ) affecting the sediment transport employing ML models (Bagging (BA), Random Committee (RC)) and the standalone ML models (Multi-Layer Perceptron Regression (MLPR) and Reduced Error Pruning Tree (REPT). The hybrid Bagging-REPT (BA-REPT) model outperformed other models with a Nash-Sutcliffe Efficiency (NSE) of 0. 915, followed by RC-REPT (NSE = 0. 906). Among the various variables, friction slope (S f ) was identified as the most influential variable affecting sediment transport behavior. It was also observed that Hybrid models can predict sediment transport behavior more accurately as compared to standalone models and empirical equations. The findings of the study thus demonstrate the importance of hybrid learning in addressing the nonlinear complexity of sediment transport processes.

Details DOI

ICLR Conference 2024 Conference Paper

An Investigation of Representation and Allocation Harms in Contrastive Learning

Subha Maity
Mayank Agarwal
Mikhail Yurochkin
Yuekai Sun

The effect of underrepresentation on the performance of minority groups is known to be a serious problem in supervised learning settings; however, it has been underexplored so far in the context of self-supervised learning (SSL). In this paper, we demonstrate that contrastive learning (CL), a popular variant of SSL, tends to collapse representations of minority groups with certain majority groups. We refer to this phenomenon as representation harm and demonstrate it on image and text datasets using the corresponding popular CL methods. Furthermore, our causal mediation analysis of allocation harm on a downstream classification task reveals that representation harm is partly responsible for it, thus emphasizing the importance of studying and mitigating representation harm. Finally, we provide a theoretical explanation for representation harm using a stochastic block model that leads to a representational neural collapse in a contrastive learning setting.

Details

EAAI Journal 2023 Journal Article

DLIRIR: Deep learning based improved Reverse Image Retrieval

Divya Singh
Jimson Mathew
Mayank Agarwal
Mahesh Govind

Deep feature search has transpired as a sub-problem of large-scale Reverse Image Retrieval (RIR) algorithms in computer vision. In light of recent progress, transforming deep features into binary codes and employing various search strategies to delineate similar images has gained popularity. However, these binary codes cannot accurately define the similarity between two images, which is essential for image retrieval. Moreover, when retrieving k closest images, the existing RIR approaches show a drop in precision value as k increases. In this paper, we put forward a method, Deep Learning based Improved Reverse Image Retrieval (DLIRIR), which solves the RIR problem by utilizing a deep neural network, clustering methodology, and dimensionality reduction technique. Motivated by the need to prevent information loss in converting deep features to binary codes and reduce search time, we cluster image features and select some features from each cluster called Representative Points (RPs). These RPs are then compared to features of an image (query) to retrieve images similar to it from the database. The Euclidean distances of the query feature and the RPs indicate similarity between the query and the cluster of images represented by that RP. We reduce the number of comparisons by selecting as few RPs as possible. We evaluate the performance of DLIRIR by comparing it with other state-of-the-art algorithms in the RIR domain. Our method achieves a precision greater than 99% in benchmark datasets Caltech 256, Corel 10k, and Adaptiope, while a precision between 80%–85% on large size datasets Cifar 100, Tiny ImageNet, and VGGFace2.

Details DOI

NeurIPS Conference 2021 Conference Paper

On sensitivity of meta-learning to support data

Mayank Agarwal
Mikhail Yurochkin
Yuekai Sun

Meta-learning algorithms are widely used for few-shot learning. For example, image recognition systems that readily adapt to unseen classes after seeing only a few labeled examples. Despite their success, we show that modern meta-learning algorithms are extremely sensitive to the data used for adaptation, i. e. support data. In particular, we demonstrate the existence of (unaltered, in-distribution, natural) images that, when used for adaptation, yield accuracy as low as 4\% or as high as 95\% on standard few-shot image classification benchmarks. We explain our empirical findings in terms of class margins, which in turn suggests that robust and safe meta-learning requires larger margins than supervised learning.

PDF Details

AAAI Conference 2020 System Paper

TraceHub – A Platform to Bridge the Gap between State-of-the-Art Time-Series Analytics and Datasets

Shubham Agarwal
Christian Muise
Mayank Agarwal
Sohini Upadhyay
Zilu Tang
Zhongshen Zeng
Yasaman Khazaeni

In this paper, we present TraceHub - a platform that connects new non-trivial state-of-the-art time-series analytics with datasets from different domains. Analytics owners can run their insights on new datasets in an automated setting to ﬁnd insight’s potential and improve it. Dataset owners can ﬁnd all possible types of non-trivial insights based on latest research. We provide a plug-n-play system as a set of Dataset, Transformer pipeline, and Analytics APIs for both kinds of users. We show a usefulness measure of generated insights across various types of analytics in the system. We believe that this platform can be used to bridge the gap between time-series analytics and datasets by signiﬁcantly reducing the time to ﬁnd the true potential of budding timeseries research and improving on it faster.

PDF Details

ICML Conference 2019 Conference Paper

Bayesian Nonparametric Federated Learning of Neural Networks

Mikhail Yurochkin
Mayank Agarwal
Soumya Ghosh
Kristjan H. Greenewald
Trong Nghia Hoang
Yasaman Khazaeni

In federated learning problems, data is scattered across different servers and exchanging or pooling it is often impractical or prohibited. We develop a Bayesian nonparametric framework for federated learning with neural networks. Each data server is assumed to provide local neural network weights, which are modeled through our framework. We then develop an inference approach that allows us to synthesize a more expressive global network without additional supervision, data pooling and with as few as a single communication round. We then demonstrate the efficacy of our approach on federated learning problems simulated from two popular image classification datasets.

Details

AAAI Conference 2019 Conference Paper

CAPNet: Continuous Approximation Projection for 3D Point Cloud Reconstruction Using 2D Supervision

K. L. Navaneet
Priyanka Mandikal
Mayank Agarwal
R. Venkatesh Babu

Knowledge of 3D properties of objects is a necessity in order to build effective computer vision systems. However, lack of large scale 3D datasets can be a major constraint for datadriven approaches in learning such properties. We consider the task of single image 3D point cloud reconstruction, and aim to utilize multiple foreground masks as our supervisory data to alleviate the need for large scale 3D datasets. A novel differentiable projection module, called ‘CAPNet’, is introduced to obtain such 2D masks from a predicted 3D point cloud. The key idea is to model the projections as a continuous approximation of the points in the point cloud. To overcome the challenges of sparse projection maps, we propose a loss formulation termed ‘affinity loss’ to generate outlierfree reconstructions. We significantly outperform the existing projection based approaches on a large-scale synthetic dataset. We show the utility and generalizability of such a 2D supervised approach through experiments on a real-world dataset, where lack of 3D data can be a serious concern. To further enhance the reconstructions, we also propose a test stage optimization procedure to obtain reconstructions that display high correspondence with the observed input image.

PDF Details

NeurIPS Conference 2019 Conference Paper

Statistical Model Aggregation via Parameter Matching

Mikhail Yurochkin
Mayank Agarwal
Soumya Ghosh
Kristjan Greenewald
Nghia Hoang

We consider the problem of aggregating models learned from sequestered, possibly heterogeneous datasets. Exploiting tools from Bayesian nonparametrics, we develop a general meta-modeling framework that learns shared global latent structures by identifying correspondences among local model parameterizations. Our proposed framework is model-independent and is applicable to a wide range of model types. After verifying our approach on simulated data, we demonstrate its utility in aggregating Gaussian topic models, hierarchical Dirichlet process based hidden Markov models, and sparse Gaussian processes with applications spanning text summarization, motion capture analysis, and temperature forecasting.

PDF Details