Search Results for author:
Found 53 papers, 15 papers with code
Programming Refusal with Conditional Activation Steering
In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context.
CELL your Model: Contrastive Explanation Methods for Large Language Models
However, in the case of generative AI such as large language models (LLMs), there is no class prediction to explain.
The global landscape of academic guidelines for generative AI and Large Language Models
The integration of Generative Artificial Intelligence (GAI) and Large Language Models (LLMs) in academia has spurred a global discourse on their potential pedagogical benefits and ethical considerations.
Deep Generative Sampling in the Dual Divergence Space: A Data-efficient & Interpretative Approach for Generative AI
In addition to the data efficiency gained from direct sampling, we propose an algorithm that offers a significant reduction in sample complexity for estimating the divergence of the data distribution with respect to the marginal distribution.
Multi-Level Explanations for Generative Language Models
To address the challenges of text as output and long text inputs, we propose a general framework called MExGen that can be instantiated with different attribution algorithms.
NeuroPrune: A Neuro-inspired Topological Sparse Training Algorithm for Large Language Models
Transformer-based Language Models have become ubiquitous in Natural Language Processing (NLP) due to their impressive performance on various tasks.
Ranking Large Language Models without Ground Truth
and a set of LLMs, we rank them without access to any ground truth or reference responses.
Trust Regions for Explanations via Black-Box Probabilistic Certification
fidelity, stability), can we find the largest hypercube (i. e., $\ell_{\infty}$ ball) centered at the example such that when the explanation is applied to all examples within the hypercube, (with high probability) a quality criterion is met (viz.
Spectral Adversarial MixUp for Few-Shot Unsupervised Domain Adaptation
To accomplish this challenging task, first, a spectral sensitivity map is introduced to characterize the generalization weaknesses of models in the frequency domain.
When Neural Networks Fail to Generalize? A Model Sensitivity Perspective
Domain generalization (DG) aims to train a model to perform well in unseen domains under different distributions.
On the Safety of Interpretable Machine Learning: A Maximum Deviation Approach
Interpretable and explainable machine learning has seen a recent surge of interest.
Reprogramming Pretrained Language Models for Antibody Sequence Infilling
Results on antibody design benchmarks show that our model on low-resourced antibody sequence dataset provides highly diverse CDR sequences, up to more than a two-fold increase of diversity over the baselines, without losing structural integrity and naturalness.
PainPoints: A Framework for Language-based Detection of Chronic Pain and Expert-Collaborative Text-Summarization
Chronic pain is a pervasive disorder which is often very disabling and is associated with comorbidities such as depression and anxiety.
Anomaly Attribution with Likelihood Compensation
In either case, one would ideally want to compute a ``responsibility score'' indicative of the extent to which an input variable is responsible for the anomalous output.
Atomist or Holist? A Diagnosis and Vision for More Productive Interdisciplinary AI Ethics Dialogue
In response to growing recognition of the social impact of new AI-based technologies, major AI and ML conferences and journals now encourage or require papers to include ethics impact statements and undergo ethics reviews.
Connecting Algorithmic Research and Usage Contexts: A Perspective of Contextualized Evaluation for Explainable AI
We argue that one way to close the gap is to develop evaluation methods that account for different user requirements in these usage contexts.
Accurate Clinical Toxicity Prediction using Multi-task Deep Neural Nets and Contrastive Molecular Explanations
Additionally, our multi-task approach is comprehensive in the sense that it is comparable to state-of-the-art approaches for specific endpoints in in vitro, in vivo and clinical platforms.
Local Explanations for Reinforcement Learning
Many works in explainable AI have focused on explaining black-box classification models.
Analogies and Feature Attributions for Model Agnostic Explanation of Similarity Learners
We first propose a method that provides feature attributions to explain the similarity between a pair of inputs as determined by a black box similarity learner.
Auto-Transfer: Learning to Route Transferrable Representations
Knowledge transfer between heterogeneous source and target networks and tasks has received a lot of attention in recent times as large amounts of quality labeled data can be difficult to obtain in many applications.
CoFrNets: Interpretable Neural Architecture Inspired by Continued Fractions
We experiment on nonlinear synthetic functions and are able to accurately model as well as estimate feature attributions and even higher order terms in some cases, which is a testament to the representational power as well as interpretability of such architectures.
Auto-Transfer: Learning to Route Transferable Representations
Knowledge transfer between heterogeneous source and target networks and tasks has received a lot of attention in recent times as large amounts of quality labelled data can be difficult to obtain in many applications.
Locally Invariant Explanations: Towards Causal Explanations through Local Invariant Learning
Locally interpretable model agnostic explanations (LIME) method is one of the most popular methods used to explain black-box models at a per example level.
Interpreting Reinforcement Policies through Local Behaviors
Many works in explainable AI have focused on explaining black-box classification models.
AI Explainability 360: Impact and Design
As artificial intelligence and machine learning algorithms become increasingly prevalent in society, multiple stakeholders are calling for these algorithms to provide explanations.
Let the CAT out of the bag: Contrastive Attributed explanations for Text
Contrastive explanations for understanding the behavior of black box models has gained a lot of attention recently as they provide potential for recourse.
Multihop: Leveraging Complex Models to Learn Accurate Simple Models
In this paper, we propose a meta-approach where we transfer information from the complex model to the simple model by dynamically selecting and/or constructing a sequence of intermediate models of decreasing complexity that are less intricate than the original complex model.
Explainable artificial intelligence
Knowledge Distillation
+2
Towards Better Model Understanding with Path-Sufficient Explanations
To overcome these limitations, we propose a novel method called Path-Sufficient Explanations Method (PSEM) that outputs a sequence of sufficient explanations for a given input of strictly decreasing size (or value) -- from original input to a minimally sufficient explanation -- which can be thought to trace the local boundary of the model in a smooth manner, thus providing better intuition about the local model behavior for the specific input.
Explainable artificial intelligence
Explainable Artificial Intelligence (XAI)
Treatment Effect Estimation using Invariant Risk Minimization
Inferring causal individual treatment effect (ITE) from observational data is a challenging problem whose difficulty is exacerbated by the presence of treatment assignment bias.
Learning to Initialize Gradient Descent Using Gradient Descent
Non-convex optimization problems are challenging to solve; the success and computational expense of a gradient descent algorithm or variant depend heavily on the initialization strategy.
Empirical or Invariant Risk Minimization? A Sample Complexity Perspective
Recently, invariant risk minimization (IRM) was proposed as a promising solution to address out-of-distribution (OOD) generalization.
Linear Regression Games: Convergence Guarantees to Approximate Out-of-Distribution Solutions
In Ahuja et al., it was shown that solving for the Nash equilibria of a new class of "ensemble-games" is equivalent to solving IRM.
Deciding Fast and Slow: The Role of Cognitive Biases in AI-assisted Decision-making
We, then, conduct a second user experiment which shows that our time allocation strategy with explanation can effectively de-anchor the human and improve collaborative performance when the AI model has low confidence and is incorrect.
Model Agnostic Multilevel Explanations
The method can also leverage side information, where users can specify points for which they may want the explanations to be similar.
Learning Global Transparent Models Consistent with Local Contrastive Explanations
Based on a key insight we propose a novel method where we create custom boolean features from sparse local contrastive explanations of the black-box model and then train a globally transparent model on just these, and showcase empirically that such models have higher local consistency compared with other known strategies, while still being close in performance to models that are trained with access to the original data.
Invariant Risk Minimization Games
The standard risk minimization paradigm of machine learning is brittle when operating in environments whose test distributions are different from the training distribution due to spurious correlations.
Leveraging Simple Model Predictions for Enhancing its Performance
Our method also leverages the per sample hardness estimate of the simple model which is not the case with the prior works which primarily consider the complex model's confidences/predictions and is thus conceptually novel.
One Explanation Does Not Fit All: A Toolkit and Taxonomy of AI Explainability Techniques
Equally important, we provide a taxonomy to help entities requiring explanations to navigate the space of explanation methods, not only those in the toolkit but also in the broader literature on explainability.
Teaching AI to Explain its Decisions Using Embeddings and Multi-Task Learning
Using machine learning in high-stakes applications often requires predictions to be accompanied by explanations comprehensible to the domain user, who has ultimate responsibility for decisions and outcomes.
Model Agnostic Contrastive Explanations for Structured Data
Recently, a method [7] was proposed to generate contrastive explanations for differentiable models such as deep neural networks, where one has complete access to the model.
Enhancing Simple Models by Exploiting What They Already Know
Our method also leverages the per sample hardness estimate of the simple model which is not the case with the prior works which primarily consider the complex model's confidences/predictions and is thus conceptually novel.
Leveraging Latent Features for Local Explanations
As the application of deep neural networks proliferates in numerous areas such as medical imaging, video surveillance, and self driving cars, the need for explaining the decisions of these models has become a hot research topic, both at the global and local level.
TED: Teaching AI to Explain its Decisions
Artificial intelligence systems are being increasingly deployed due to their potential to increase the efficiency, scale, consistency, fairness, and accuracy of decisions.
Streaming Methods for Restricted Strongly Convex Functions with Applications to Prototype Selection
In this paper, we show that if the optimization function is restricted-strongly-convex (RSC) and restricted-smooth (RSM) -- a rich subclass of weakly submodular functions -- then a streaming algorithm with constant factor approximation guarantee is possible.
Improving Simple Models with Confidence Profiles
Our transfer method involves a theoretically justified weighting of samples during the training of the simple model using confidence scores of these intermediate layers.
Teaching Meaningful Explanations
The adoption of machine learning in high-stakes applications such as healthcare and law has lagged in part because predictions are not accompanied by explanations comprehensible to the domain user, who often holds the ultimate responsibility for decisions and outcomes.
Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negatives
important object pixels in an image) to justify its classification and analogously what should be minimally and necessarily \emph{absent} (viz.
A Formal Framework to Characterize Interpretability of Procedures
We provide a novel notion of what it means to be interpretable, looking past the usual association with human understanding.
Efficient Data Representation by Selecting Prototypes with Importance Weights
Prototypical examples that best summarizes and compactly represents an underlying complex data distribution communicate meaningful insights to humans in domains where simple explanations are hard to extract.
TIP: Typifying the Interpretability of Procedures
This leads to the insight that the improvement in the target model is not only a function of the oracle model's performance, but also its relative complexity with respect to the target model.
Learning with Changing Features
In particular, we propose an approach to provably determine the time instant from which the new/changed features start becoming relevant with respect to an output variable in an agnostic (supervised) learning setting.
Uncovering Group Level Insights with Accordant Clustering
Clustering is a widely-used data mining tool, which aims to discover partitions of similar items in data.
Building an Interpretable Recommender via Loss-Preserving Transformation
We propose a method for building an interpretable recommender system for personalizing online content and promotions.
