Author name cluster

Venkatesh Babu Radhakrishnan

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.

9 papers

1 author row

AAAI Conference 2026 Conference Paper

Harnessing Diffusion-Generated Synthetic Images for Fair Image Classification

Abhipsa Basu
Aviral Gupta
Abhijnya Bhat
Venkatesh Babu Radhakrishnan

Image classification systems often inherit biases from uneven group representation in training data. For example, in face datasets for hair color classification, blond hair may be disproportionately associated with females, reinforcing stereotypes. A recent approach leverages the Stable Diffusion model to generate balanced training data, but these models often struggle to preserve the original data distribution. In this work, we explore multiple diffusion-finetuning techniques, e.g., LoRA and DreamBooth, to generate images that more accurately represent each training group by learning directly from their samples. Additionally, in order to prevent a single DreamBooth model from being overwhelmed by excessive intra-group variations, we explore a technique of clustering images within each group and train a DreamBooth model per cluster. These models are then used to generate group-balanced data for pretraining, followed by fine-tuning on real data. Experiments on multiple benchmarks demonstrate that the studied finetuning approaches outperform vanilla Stable Diffusion on average and achieve results comparable to SOTA debiasing techniques like Group-DRO, while surpassing them as the dataset bias severity increases.

PDF Details DOI

TMLR Journal 2026 Journal Article

Interpreting Kolmogorov-Arnold Networks in Neuroimaging: A Path-Based Attribution Framework

Suhrud Murthy
Venkatesh Babu Radhakrishnan
Neelam Sinha

Explainability aspects of most classification models are learnt through instance-specific analysis. However, in understanding diseases, it is important to consider population-wide analysis in order to identify affected regions that are consistently seen across cohorts of diseased population. In this study, we report utility of Kolmogorov-Arnold Networks (KANs) in understanding population-wide characteristics seen in subjects affected by Alzheimer's disease (AD). KANs offer enhanced interpretability through learnable activation functions on network edges. Thus, the learned functions reflect the characteristics of the entire span of training data. In a KAN network trained for classification, attributions through the network can be traced to understand how specific inputs influence the output label. In this study, we propose a path-based attribution framework that generates global importance maps by tracing exhaustive information flow through all potential paths. Our method initially scores the functions on the edges of a trained KAN using an appropriate scoring function. Subsequently, these scores are propagated through the network to compute path-attributions. This approach scales linearly with network depth, and is only dependent on model training and does not need further analysis on training data post-hoc. Evaluation on three public AD neuroimaging datasets (OASIS, ADNI, Mendeley, totally comprising 7428 acquisitions), were carried out on 2D brain slices as well as 3D brain volumes. The corresponding KAN test accuracies are $93.24\%$, $81.85\%$, and $91.25\%$ on OASIS, ADNI, and Mendeley datasets, respectively. Alongside, competitive or improved performance via metrics such as Insertion AUC, Deletion AUC and Sufficiency, is also demonstrated. The generated attribution maps identify clinically meaningful regions including the body and genu of corpus callossum, corona radiata, bilateral caudate nuclei, medial prefrontal cortex and temporal lobe structures, aligned with established AD pathology literature. By providing voxel-level global attributions as network-intrinsic properties, our framework addresses a critical gap in AI interpretability and supports exploratory clinical analysis and model auditing of AI-assisted AD diagnosis systems.

PDF Details

AAAI Conference 2026 Conference Paper

UniC-Lift: Unified 3D Instance Segmentation via Contrastive Learning

Ankit Dhiman
Srinath R
Jaswanth Reddy
Lokesh R Boregowda
Venkatesh Babu Radhakrishnan

3D Gaussian Splatting (3DGS) and Neural Radiance Fields (NeRF) have advanced novel-view synthesis. Recent methods extend multi-view 2D segmentation to 3D, enabling instance/semantic segmentation for better scene understanding. A key challenge is the inconsistency of 2D instance labels across views, leading to poor 3D predictions. Existing methods use a two-stage approach in which some rely on contrastive learning with hyperparameter-sensitive clustering, while others preprocess labels for consistency. We propose a unified framework that merges these steps, reducing training time and improving performance by introducing a learnable feature embedding for segmentation in Gaussian primitives. This embedding is then efficiently decoded into instance labels through a novel "Embedding-to-Label" process, effectively integrating the optimization. While this unified framework offers substantial benefits, we observed artifacts at the object boundaries. To address the object boundary issues, we propose hard-mining samples along these boundaries. However, directly applying hard mining to the feature embeddings proved unstable. Therefore, we apply a linear layer to the rasterized feature embeddings before calculating the triplet loss, which stabilizes training and significantly improves performance. Our method outperforms baselines qualitatively and quantitatively on the ScanNet, Replica3D, and Messy-Rooms datasets.

PDF Details DOI

TMLR Journal 2025 Journal Article

Concept Siever: Towards Controllable Erasure of Concepts from Diffusion Models without Side-effect

Aakash Kumar Singh
Priyam Dey
Sribhav Srivatsa
Venkatesh Babu Radhakrishnan

Diffusion models' unprecedented success with image generation can largely be attributed to their large-scale pretraining on massive datasets. Yet, the necessity of forgetting specific concepts for regulatory or copyright compliance poses a critical challenge. Existing approaches in concept forgetting, although reasonably successful in forgetting a given concept, frequently fail to preserve generation quality or demand extensive domain expertise for preservation. To alleviate such issues, we introduce Concept Siever, an end-to-end framework for targeted concept removal within pre-trained text-to-image diffusion models. The foundation of Concept Siever rests on \textit{two key innovations}: First, an automatic technique to create paired dataset of target concept and its negations by utilizing the diffusion model’s latent space. A key property of these pairs is that they differ only in the target concept, enabling forgetting with \textit{minimal side effects} and \textit{without requiring domain expertise}. Second, we present Concept Sieve, a localization method for identifying and isolating the model components most responsible to the target concept. By retraining only these localized components on our paired dataset for a target concept, Concept Siever accurately removes the concept with \textit{negligible side-effects, preserving neighboring and unrelated concepts}. Moreover, given the subjective nature of forgetting a concept like nudity, we propose Concept Sieve which provides a \texit{fine-grained control over the forgetting strength at inference time}, catering to diverse deployment needs without any need of finetuning. We report state-of-the-art performance on the I2P benchmark, surpassing previous domain-agnostic methods by over $33\%$ while showing superior structure preservation. We validate our results through extensive quantitative and qualitative evaluation along with a user study.

PDF Details

TMLR Journal 2024 Journal Article

ProFeAT: Projected Feature Adversarial Training for Self-Supervised Learning of Robust Representations

Sravanti Addepalli
Priyam Dey
Venkatesh Babu Radhakrishnan

The need for abundant labelled data in supervised Adversarial Training (AT) has prompted the use of Self-Supervised Learning (SSL) techniques with AT. However, the direct application of existing SSL methods to adversarial training has been sub-optimal due to the increased training complexity of combining SSL with AT. A recent approach DeACL \citep{deacl} mitigates this by utilizing supervision from a standard SSL teacher in a distillation setting, to mimic supervised AT. However, we find that there is still a large performance gap when compared to supervised adversarial training, specifically on larger models. In this work, investigate the key reason for this gap and propose Projected Feature Adversarial Training (ProFeAT) to bridge the same. We show that the sub-optimal distillation performance is a result of mismatch in training objectives of the teacher and student, and propose to use a projection head at the student, that allows it to leverage weak supervision from the teacher while also being able to learn adversarially robust representations that are distinct from the teacher. We further propose appropriate attack and defense losses at the feature and projector, alongside a combination of weak and strong augmentations for the teacher and student respectively, to improve the training data diversity without increasing the training complexity. Through extensive experiments on several benchmark datasets and models, we demonstrate significant improvements in both clean and robust accuracy when compared to existing SSL-AT methods, setting a new state-of-the-art. We further report on-par/ improved performance when compared to TRADES, a popular supervised-AT method.

PDF Details

AAAI Conference 2022 Conference Paper

Amplitude Spectrum Transformation for Open Compound Domain Adaptive Semantic Segmentation

Jogendra Nath Kundu
Akshay R Kulkarni
Suvaansh Bhambri
Varun Jampani
Venkatesh Babu Radhakrishnan

Open compound domain adaptation (OCDA) has emerged as a practical adaptation setting which considers a single labeled source domain against a compound of multi-modal unlabeled target data in order to generalize better on novel unseen domains. We hypothesize that an improved disentanglement of domain-related and task-related factors of dense intermediate layer features can greatly aid OCDA. Prior-arts attempt this indirectly by employing adversarial domain discriminators on the spatial CNN output. However, we find that latent features derived from the Fourier-based amplitude spectrum of deep CNN features hold a more tractable mapping with domain discrimination. Motivated by this, we propose a novel feature space Amplitude Spectrum Transformation (AST). During adaptation, we employ the AST auto-encoder for two purposes. First, carefully mined source-target instance pairs undergo a simulation of cross-domain feature stylization (AST- Sim) at a particular layer by altering the AST-latent. Second, AST operating at a later layer is tasked to normalize (AST-Norm) the domain content by fixing its latent to a mean prototype. Our simplified adaptation technique is not only clustering-free but also free from complex adversarial alignment. We achieve leading performance against the prior arts on the OCDA scene segmentation benchmarks.

PDF Details

AAAI Conference 2022 Conference Paper

Beyond Learning Features: Training a Fully-Functional Classifier with ZERO Instance-Level Labels

Deepak Babu Sam
Abhinav Agarwalla
Venkatesh Babu Radhakrishnan

We attempt to train deep neural networks for classification without using any labeled data. Existing unsupervised methods, though mine useful clusters or features, require some annotated samples to facilitate the final task-specific predictions. This defeats the true purpose of unsupervised learning and hence we envisage a paradigm of ‘true’ self-supervision, where absolutely no annotated instances are used for training a classifier. The proposed method first pretrains a deep network through self-supervision and performs clustering on the learned features. A classifier layer is then appended to the self-supervised network and is trained by matching the distribution of the predictions to that of a predefined prior. This approach leverages the distribution of labels for supervisory signals and consequently, no image-label pair is needed. Experiments reveal that the method works on major nominal as well as ordinal classification datasets and delivers significant performance.

PDF Details

AAAI Conference 2020 Conference Paper

DeGAN: Data-Enriching GAN for Retrieving Representative Samples from a Trained Classifier

Sravanti Addepalli
Gaurav Kumar Nayak
Anirban Chakraborty
Venkatesh Babu Radhakrishnan

In this era of digital information explosion, an abundance of data from numerous modalities is being generated as well as archived everyday. However, most problems associated with training Deep Neural Networks still revolve around lack of data that is rich enough for a given task. Data is required not only for training an initial model, but also for future learning tasks such as Model Compression and Incremental Learning. A diverse dataset may be used for training an initial model, but it may not be feasible to store it throughout the product life cycle due to data privacy issues or memory constraints. We propose to bridge the gap between the abundance of available data and lack of relevant data, for the future learning tasks of a given trained network. We use the available data, that may be an imbalanced subset of the original training dataset, or a related domain dataset, to retrieve representative samples from a trained classiﬁer, using a novel Dataenriching GAN (DeGAN) framework. We demonstrate that data from a related domain can be leveraged to achieve stateof-the-art performance for the tasks of Data-free Knowledge Distillation and Incremental Learning on benchmark datasets. We further demonstrate that our proposed framework can enrich any data, even from unrelated domains, to make it more useful for the future learning tasks of a given network.

PDF Details

AAAI Conference 2020 Conference Paper

Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation

Jogendra Nath Kundu
Siddharth Seth
Rahul M V
Mugalodi Rakesh
Venkatesh Babu Radhakrishnan
Anirban Chakraborty

Estimation of 3D human pose from monocular image has gained considerable attention, as a key step to several humancentric applications. However, generalizability of human pose estimation models developed using supervision on large-scale in-studio datasets remains questionable, as these models often perform unsatisfactorily on unseen in-the-wild environments. Though weakly-supervised models have been proposed to address this shortcoming, performance of such models relies on availability of paired supervision on some related task, such as 2D pose or multi-view image pairs. In contrast, we propose a novel kinematic-structure-preserved unsupervised 3D pose estimation framework, which is not restrained by any paired or unpaired weak supervisions. Our pose estimation framework relies on a minimal set of prior knowledge that deﬁnes the underlying kinematic 3D structure, such as skeletal joint connectivity information with bone-length ratios in a ﬁxed canonical scale. The proposed model employs three consecutive differentiable transformations namely forwardkinematics, camera-projection and spatial-map transformation. This design not only acts as a suitable bottleneck stimulating effective pose disentanglement, but also yields interpretable latent pose representations avoiding training of an explicit latent embedding to pose mapper. Furthermore, devoid of unstable adversarial setup, we re-utilize the decoder to formalize an energy-based loss, which enables us to learn from in-the-wild videos, beyond laboratory settings. Comprehensive experiments demonstrate our state-of-the-art unsupervised and weakly-supervised pose estimation performance on both Human3. 6M and MPI-INF-3DHP datasets. Qualitative results on unseen environments further establish our superior generalization ability.

PDF Details