Covariate shift in the test data is a common practical phenomena that can significantly downgrade both the accuracy and the fairness performance of the model. Ensuring fairness across different sensitive groups under covariate shift is of paramount importance due to societal implications like criminal justice. We operate in the unsupervised regime where only a small set of unlabeled test samples along with a labeled training set is available. Towards improving fairness under this highly challenging yet realistic scenario, we make three contributions. First is a novel composite weighted entropy based objective for prediction accuracy which is optimized along with a representation matching loss for fairness. We experimentally verify that optimizing with our loss formulation outperforms a number of state-of-the-art baselines in the pareto sense with respect to the fairness-accuracy tradeoff on several standard datasets. Our second contribution is a new setting we term Asymmetric Covariate Shift that, to the best of our knowledge, has not been studied before. Asymmetric covariate shift occurs when distribution of covariates of one group shifts significantly compared to the other groups and this happens when a dominant group is over-represented. While this setting is extremely challenging for current baselines, We show that our proposed method significantly outperforms them. Our third contribution is theoretical, where we show that our weighted entropy term along with prediction loss on the training set approximates test loss under covariate shift. Empirically and through formal sample complexity bounds, we show that this approximation to the unseen test loss does not depend on importance sampling variance which affects many other baselines.

Molecular Representation Learning (MRL) has proven impactful in numerous biochemical applications such as drug discovery and enzyme design. While Graph Neural Networks (GNNs) are effective at learning molecular representations from a 2D molecular graph or a single 3D structure, existing works often overlook the flexible nature of molecules, which continuously interconvert across conformations via chemical bond rotations and minor vibrational perturbations. To better account for molecular flexibility, some recent works formulate MRL as an ensemble learning problem, focusing on explicitly learning from a set of conformer structures. However, most of these studies have limited datasets, tasks, and models. In this work, we introduce the first MoleculAR Conformer Ensemble Learning (MARCEL) benchmark to thoroughly evaluate the potential of learning on con- former ensembles and suggest promising research directions. MARCEL includes four datasets covering diverse molecule- and reaction-level properties of chemically diverse molecules including organocatalysts and transition-metal catalysts, extending beyond the scope of common GNN benchmarks that are confined to drug-like molecules. In addition, we conduct a comprehensive empirical study, which benchmarks representative 1D, 2D, and 3D MRL models, along with two strategies that explicitly incorporate conformer ensembles into 3D models. Our findings reveal that direct learning from an accessible conformer space can improve performance on a variety of tasks and models.

We study the problem of adversarial attack and robustness on tabular datasets with discrete features. The discrete features of a tabular dataset represent high-level meaningful concepts, with different sets of vocabularies, leading to requiring non-uniform robustness. Further, the notion of distance between tabular input instances is not well defined, making the problem of producing adversarial examples with minor perturbations qualitatively more challenging compared to existing methods. Towards this, our paper defines the notion of distance through the lens of feature embeddings, learnt to represent the discrete features. We then formulate the task of generating adversarial examples as abinary set selection problem under non-uniform feature importance. Next, we propose an efficient approximate gradient-descent based algorithm, calledDiscrete Non-uniform Approximation (DNA) attack, by reformulating the problem into a continuous domain to solve the original optimization problem for generating adversarial examples. We demonstrate the effectiveness of our proposed DNA attack using two large real-world discrete tabular datasets from e-commerce domains for binary classification, where the datasets are heavily biased for one-class. We also analyze challenges for existing adversarial training frameworks for such datasets under our DNA attack.

Despite the demonstrated empirical efficacy of prompt tuning to adapt a pretrained language model for a new task, the theoretical underpinnings of the difference between "tuning parameters before the input" against "the tuning of model weights" are limited. We thus take one of the first steps to understand the role of soft-prompt tuning for transformer-based architectures. By considering a general purpose architecture, we analyze prompt tuning from the lens of both: universal approximation and limitations with finite-depth fixed-weight pretrained transformers for continuous-valued functions. Our universality result guarantees the existence of a strong transformer with a prompt to approximate any sequence-to-sequence function in the set of Lipschitz functions. The limitations of prompt tuning for limited-depth transformers are first proved by constructing a set of datasets, that cannot be memorized by a prompt of any length for a given single encoder layer. We also provide a lower bound on the required number of tunable prompt parameters and compare the result with the number of parameters required for a low-rank update (based on LoRA) for a single-layer setting. We finally extend our analysis to multi-layer settings by providing sufficient conditions under which the transformer can at best learn datasets from invertible functions only. Our theoretical claims are also corroborated by empirical results.

We study the weak supervision learning problem of Learning from Label Proportions (LLP) where the goal is to learn an instance-level classifier using proportions of various class labels in a bag -- a collection of input instances that often can be highly correlated. While representation learning for weakly-supervised tasks is found to be effective, they often require domain knowledge. To the best of our knowledge, representation learning for tabular data (unstructured data containing both continuous and categorical features) are not studied. In this paper, we propose to learn diverse representations of instances within the same bags to effectively utilize the weak bag-level supervision. We propose a domain agnostic LLP method, called "Self Contrastive Representation Learning for LLP" (SelfCLR-LLP) that incorporates a novel self--contrastive function as an auxiliary loss to learn representations on tabular data for LLP. We show that diverse representations for instances within the same bags aid efficient usage of the weak bag-level LLP supervision. We evaluate the proposed method through extensive experiments on real-world LLP datasets from e-commerce applications to demonstrate the effectiveness of our proposed SelfCLR-LLP. In this paper, we propose to learn diverse representations of instances within the same bags to effectively utilize the weak bag-level supervision. We propose a domain agnostic LLP method, called "Self Contrastive Representation Learning for LLP" (SelfCLR-LLP) that incorporates a novel self--contrastive function as an auxiliary loss to learn representations on tabular data for LLP. We show that diverse representations for instances within the same bags aid efficient usage of the weak bag-level LLP supervision. We evaluate the proposed method through extensive experiments on real-world LLP datasets from e-commerce applications to demonstrate the effectiveness of our proposed SelfCLR-LLP.

Proposing scoring functions to effectively understand, analyze and learn various properties of high dimensional hidden representations of large-scale transformer models like BERT can be a challenging task. In this work, we explore a new direction by studying the topological features of BERT hidden representations using persistent homology (PH). We propose a novel scoring function named “persistence scoring function (PSF)” which: (i) accurately captures the homology of the high-dimensional hidden representations and correlates well with the test set accuracy of a wide range of datasets and outperforms existing scoring metrics, (ii) captures interesting post fine-tuning “per-class” level properties from both qualitative and quantitative viewpoints, (iii) is more stable to perturbations as compared to the baseline functions, which makes it a very robust proxy, and (iv) finally, also serves as a predictor of the attack success rates for a wide category of black-box and white-box adversarial attack methods. Our extensive correlation experiments demonstrate the practical utility of PSF on various NLP tasks relevant to BERT11Code is available at https://github.com/chauhanjatin10/BERTops

We formulate a new inference task in the domain of multivariate time series forecasting (MTSF), called Variable Subset Forecast (VSF), where only a small subset of the variables is available during inference. Variables are absent during inference because of long-term data loss (eg. sensor failures) or high -> low-resource domain shift between train / test. To the best of our knowledge, robustness of MTSF models in presence of such failures, has not been studied in the literature. Through extensive evaluation, we first show that the performance of state of the art methods degrade significantly in the VSF setting. We propose a non-parametric, wrapper technique that can be applied on top any existing forecast models. Through systematic experiments across 4 datasets and 5 forecast models, we show that our technique is able to recover close to 95% performance of the models even when only 15% of the original variables are present.

Several state-of-the-art neural graph embedding methods are based on short random walks (stochastic processes) because of their ease of computation, simplicity in capturing complex local graph properties, scalability, and interpretibility. In this work, we are interested in studying how much a probabilistic bias in this stochastic process affects the quality of the nodes picked by the process. In particular, our biased walk, with a certain probability, favors movement towards nodes whose neighborhoods bear a structural resemblance to the current node's neighborhood. We succinctly capture this neighborhood as a probability measure based on the spectrum of the node's neighborhood subgraph represented as a normalized Laplacian matrix. We propose the use of a paragraph vector model with a novel Wasserstein regularization term. We empirically evaluate our approach against several state-of-the-art node embedding techniques on a wide variety of real-world datasets and demonstrate that our proposed method significantly improves upon existing methods on both link prediction and node classification tasks.

The recent proliferation of knowledge graphs (KGs) coupled with incomplete or partial information, in the form of missing relations (links) between entities, has fueled a lot of research on knowledge base completion (also known as relation prediction).Several recent works suggest that convolutional neural network (CNN) based models generate richer and more expressive feature embeddings and hence also perform well on relation prediction.However, we observe that these KG embeddings treat triples independently and thus fail to cover the complex and hidden information that is inherently implicit in the local neighborhood surrounding a triple.To this effect, our paper proposes a novel attention-based feature embedding that captures both entity and relation features in any given entity's neighborhood.Additionally, we also encapsulate relation clusters and multi-hop relations in our model.Our empirical study offers insights into the efficacy of our attention-based model and we show marked performance gains in comparison to state-of-the-art methods on all datasets.