Search Results for author:
Found 26 papers, 8 papers with code
Multivariate Stochastic Dominance via Optimal Transport and Applications to Models Benchmarking
Our multivariate stochastic dominance test allows us to capture the dependencies between the metrics in order to make an informed and statistically significant decision on the relative performance of the models.
Distributional Preference Alignment of LLMs via Optimal Transport
Thanks to the one-dimensional nature of the resulting optimal transport problem and the convexity of the cost, it has a closed-form solution via sorting on empirical measures.
Outline-Guided Object Inpainting with Diffusion Models
We achieve that by creating variations of the available annotated object instances in a way that preserves the provided mask annotations, thereby resulting in new image-mask pairs to be added to the set of annotated images.
On the generalization capacity of neural networks during generic multimodal reasoning
On the other hand, neither of these architectural features led to productive generalization, suggesting fundamental limitations of existing architectures for specific types of multimodal generalization.
Unraveling the Key Components of OOD Generalization via Diversification
We show that (1) diversification methods are highly sensitive to the distribution of the unlabeled data used for diversification and can underperform significantly when away from a method-specific sweet spot.
Risk Aware Benchmarking of Large Language Models
Using this framework, we formally develop a risk-aware approach for foundation model selection given guardrails quantified by specified metrics.
DARE: Towards Robust Text Explanations in Biomedical and Healthcare Applications
This results in our DomainAdaptiveAREstimator (DARE) attribution robustness estimator, allowing us to properly characterize the domain-specific robustness of faithful explanations.
Adaptive Conformal Regression with Jackknife+ Rescaled Scores
We address this with a new adaptive method based on rescaling conformal scores with an estimate of local score distribution, inspired by the Jackknife+ method, which enables the use of calibration data in conformal scores without breaking calibration-test exchangeability.
Auditing and Generating Synthetic Data with Controllable Trust Trade-offs
We introduce a holistic auditing framework that comprehensively evaluates synthetic datasets and AI models.
Estimating the Adversarial Robustness of Attributions in Text with Transformers
Finally, with experiments on several text classification architectures, we show that TEA consistently outperforms current state-of-the-art AR estimators, yielding perturbations that alter explanations to a greater extent while being more fluent and less perceptible.
Active Learning for Imbalanced Civil Infrastructure Data
Unfortunately, annotation costs are incredibly high as our proprietary civil engineering dataset must be annotated by highly trained engineers.
Model-Assisted Labeling via Explainability for Visual Inspection of Civil Infrastructures
These are used within an assisted labeling framework where the annotators can interact with them as proposal segmentation masks by deciding to accept, reject or modify them, and interactions are logged as weak labels to further refine the classifier.
Compositional generalization through abstract representations in human and artificial neural networks
Our findings give empirical support to the role of compositional generalization in human behavior, implicate abstract representations as its neural implementation, and illustrate that these representations can be embedded into ANNs by designing simple and efficient pretraining procedures.
Cloud-Based Real-Time Molecular Screening Platform with MolFormer
With the prospect of automating a number of chemical tasks with high fidelity, chemical language processing models are emerging at a rapid speed.
Attention-based Interpretability with Concept Transformers
In particular, we design the Concept Transformer, a deep learning module that exposes explanations of the output of a model in which it is embedded in terms of attention over user-defined high-level concepts.
$\sbf{\delta^2}$-exploration for Reinforcement Learning
Effectively tackling the \emph{exploration-exploitation dilemma} is still a major challenge in reinforcement learning.
Deep Bandits Show-Off: Simple and Efficient Exploration with Deep Networks
Bayesian exploration strategies like Thompson Sampling resolve this trade-off in a principled way by modeling and updating the distribution of the parameters of the action-value function, the outcome model of the environment.
Image Captioning as an Assistive Technology: Lessons Learned from VizWiz 2020 Challenge
Image captioning has recently demonstrated impressive progress largely owing to the introduction of neural network algorithms trained on curated dataset like MS-COCO.
Alleviating Noisy Data in Image Captioning with Cooperative Distillation
Image captioning systems have made substantial progress, largely due to the availability of curated datasets like Microsoft COCO or Vizwiz that have accurate descriptions of their corresponding images.
Self-correcting Q-Learning
The Q-learning algorithm is known to be affected by the maximization bias, i. e. the systematic overestimation of action values, an important issue that has recently received renewed attention.
Tabular Transformers for Modeling Multivariate Time Series
This results in two architectures for tabular time series: one for learning representations that is analogous to BERT and can be pre-trained end-to-end and used in downstream tasks, and one that is akin to GPT and can be used for generation of realistic synthetic tabular sequences.
Unbalanced Sobolev Descent
USD transports particles along gradient flows of the witness function of the Sobolev-Fisher discrepancy (advection step) and reweighs the mass of particles with respect to this witness function (reaction step).
Sobolev Independence Criterion
In the kernel version we show that SIC can be cast as a convex optimization problem by introducing auxiliary variables that play an important role in feature selection as they are normalized feature importance scores.
Efficient ConvNets for Analog Arrays
Analog arrays are a promising upcoming hardware technology with the potential to drastically speed up deep learning.
Beyond Backprop: Online Alternating Minimization with Auxiliary Variables
Despite significant recent advances in deep neural networks, training them remains a challenge due to the highly non-convex nature of the objective function.
Energy-efficient neuromorphic classifiers
We also show that the energy consumption of the IBM chip is typically 2 or more orders of magnitude lower than that of conventional digital machines when implementing classifiers with comparable performance.
