Search Results for author:
Found 121 papers, 52 papers with code
Estimation of Concept Explanations Should be Uncertainty Aware
We believe the improved quality of uncertainty-aware concept explanations make them a strong candidate for more reliable model interpretation.
Parameter-Efficient Orthogonal Finetuning via Butterfly Factorization
We apply this parameterization to OFT, creating a novel parameter-efficient finetuning method, called Orthogonal Butterfly (BOFT).
Amortised Inference in Neural Networks for Small-Scale Probabilistic Meta-Learning
The global inducing point variational approximation for BNNs is based on using a set of inducing inputs to construct a series of conditional distributions that accurately approximate the conditionals of the true posterior distribution.
AI for Mathematics: A Cognitive Science Perspective
Mathematics is one of the most powerful conceptual systems developed and used by the human species.
Getting aligned on representational alignment
Finally, we lay out open problems in representational alignment where progress can benefit all three of these fields.
Pairwise Similarity Learning is SimPLE
In this paper, we focus on a general yet important learning problem, pairwise similarity learning (PSL).
Repelling Random Walks
We present a novel quasi-Monte Carlo mechanism to improve graph-based sampling, coined repelling random walks.
Universal Graph Random Features
This includes many of the most popular examples of kernels defined on the nodes of a graph.
Learning to Receive Help: Intervention-Aware Concept Embedding Models
To address this, we propose Intervention-aware Concept Embedding models (IntCEMs), a novel CBM-based architecture and training paradigm that improves a model's receptiveness to test-time interventions.
Identifying and Mitigating Privacy Risks Stemming from Language Models: A Survey
To help researchers and policymakers understand the state of knowledge around privacy attacks and mitigations, including where more work is needed, we present the first technical survey on LM privacy.
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Our MetaMath-7B model achieves 66. 4% on GSM8K and 19. 4% on MATH, exceeding the state-of-the-art models of the same size by 11. 5% and 8. 7%.
Ranked #53 on
Arithmetic Reasoning
on GSM8K
(using extra training data)
FeedbackLogs: Recording and Incorporating Stakeholder Feedback into Machine Learning Pipelines
We propose FeedbackLogs, addenda to existing documentation of ML pipelines, to track the input of multiple stakeholders.
The Future of Fundamental Science Led by Generative Closed-Loop Artificial Intelligence
Yet, AI has contributed less to fundamental science in part because large data sets of high-quality data for scientific practice and model discovery are more difficult to access.
The Neuro-Symbolic Inverse Planning Engine (NIPE): Modeling Probabilistic Social Inferences from Linguistic Inputs
To test our model, we design and run a human experiment on a linguistic goal inference task.
Selective Concept Models: Permitting Stakeholder Customisation at Test-Time
Concept-based models perform prediction using a set of concepts that are interpretable to stakeholders.
Controlling Text-to-Image Diffusion by Orthogonal Finetuning
To tackle this challenge, we introduce a principled finetuning method -- Orthogonal Finetuning (OFT), for adapting text-to-image diffusion models to downstream tasks.
Evaluating Language Models for Mathematics through Interactions
There is much excitement about the opportunity to harness the power of large language models (LLMs) when building problem-solving assistants.
Learning Personalized Decision Support Policies
In this work, we propose learning a decision support policy that, for a given input, chooses which form of support, if any, to provide.
Human Uncertainty in Concept-Based AI Systems
We study how existing concept-based models deal with uncertain interventions from humans using two novel datasets: UMNIST, a visual dataset with controlled simulated uncertainty based on the MNIST dataset, and CUB-S, a relabeling of the popular CUB concept dataset with rich, densely-annotated soft labels from humans.
Generalizing and Decoupling Neural Collapse via Hyperspherical Uniformity Gap
We then use hyperspherical uniformity (which characterizes the degree of uniformity on the unit hypersphere) as a unified framework to quantify these two objectives.
Use Perturbations when Learning from Explanations
Machine learning from explanations (MLX) is an approach to learning that uses human-provided explanations of relevant or irrelevant features for each input to ensure that model predictions are right for the right reasons.
Scalable Infomin Learning
The task of infomin learning aims to learn a representation with high utility while being uninformative about a specified target, with the latter achieved by minimising the mutual information between the representation and the target.
Optimising Human-Machine Collaboration for Efficient High-Precision Information Extraction from Text Documents
We argue for the utility of a human-in-the-loop approach in applications where high precision is required, but purely manual extraction is infeasible.
Learning a Fourier Transform for Linear Relative Positional Encodings in Transformers
We propose a new class of linear Transformers called FourierLearner-Transformers (FLTs), which incorporate a wide range of relative positional encoding mechanisms (RPEs).
Efficient Graph Field Integrators Meet Point Clouds
We present two new classes of algorithms for efficient field integration on graphs encoding point clouds.
FAVOR#: Sharp Attention Kernel Approximations via New Classes of Positive Random Features
The problem of efficient approximation of a linear operator induced by the Gaussian or softmax kernel is often addressed using random features (RFs) which yield an unbiased approximation of the operator's result.
Simplex Random Features
We present Simplex Random Features (SimRFs), a new random feature (RF) mechanism for unbiased approximation of the softmax and Gaussian kernels by geometrical correlation of random projection vectors.
Towards Robust Metrics for Concept Representation Evaluation
In this paper, we show that such metrics are not appropriate for concept learning and propose novel metrics for evaluating the purity of concept representations in both approaches.
Robust Explanation Constraints for Neural Networks
Post-hoc explanation methods are used with the intent of providing insights about neural networks and are sometimes said to help engender trust in their outputs.
Towards More Robust Interpretation via Local Gradient Alignment
However, the lack of considering the normalization of the attributions, which is essential in their visualizations, has been an obstacle to understanding and improving the robustness of feature attribution methods.
On the Informativeness of Supervision Signals
Supervised learning typically focuses on learning transferable representations from training examples annotated by humans.
Human-in-the-Loop Mixup
We focus on the synthetic data used in mixup: a powerful regularizer shown to improve model robustness, generalization, and calibration.
Iterative Teaching by Data Hallucination
We consider the problem of iterative machine teaching, where a teacher sequentially provides examples based on the status of a learner under a discrete input space (i. e., a pool of finite samples), which greatly limits the teacher's capability.
Continual Learning by Modeling Intra-Class Variation
It has been observed that neural networks perform poorly when the data or tasks are presented sequentially.
Can We Automate the Analysis of Online Child Sexual Exploitation Discourse?
Social media's growing popularity raises concerns around children's online safety.
Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Off
Deploying AI-powered systems requires trustworthy models supporting effective human interactions, going beyond raw prediction accuracy.
Structural Causal 3D Reconstruction
This paper considers the problem of unsupervised 3D object reconstruction from in-the-wild single-view images.
Eliciting and Learning with Soft Labels from Every Annotator
Our elicitation methodology therefore shows nuanced promise in enabling practitioners to enjoy the benefits of improved model performance and reliability with fewer annotators, and serves as a guide for future dataset curators on the benefits of leveraging richer information, such as categorical uncertainty, from individual annotators.
Measuring Representational Robustness of Neural Networks Through Shared Invariances
Our work offers a new view on robustness by using another reference NN to define the set of perturbations a given NN should be invariant to, thus generalizing the reliance on a reference ``human NN'' to any NN.
Chefs' Random Tables: Non-Trigonometric Random Features
We introduce chefs' random tables (CRTs), a new class of non-trigonometric random features (RFs) to approximate Gaussian and softmax kernels.
Multi-disciplinary fairness considerations in machine learning for clinical trials
While interest in the application of machine learning to improve healthcare has grown tremendously in recent years, a number of barriers prevent deployment in medical practice.
Perspectives on Incorporating Expert Feedback into Model Updates
A practitioner may receive feedback from an expert at the observation- or domain-level, and convert this feedback into updates to the dataset, loss function, or parameter space.
Synthetic Data -- what, why and how?
This explainer document aims to provide an overview of the current state of the rapidly expanding work on synthetic data technologies, with a particular focus on privacy.
On the Utility of Prediction Sets in Human-AI Teams
We explore how such prediction sets impact expert decision-making in human-AI teams.
Partitioned Variational Inference: A Framework for Probabilistic Federated Learning
Variational inference (VI) has become the method of choice for fitting many modern probabilistic models.
Robust Learning from Observation with Model Misspecification
Imitation learning (IL) is a popular paradigm for training policies in robotic systems when specifying the reward function is difficult.
Approximating Full Conformal Prediction at Scale via Influence Functions
We prove that our method is a consistent approximation of full CP, and empirically show that the approximation error becomes smaller as the training set increases; e. g., for $10^{3}$ training points the two methods output p-values that are $<10^{-3}$ apart: a negligible error for any practical application.
Filling gaps in trustworthy development of AI
The range of application of artificial intelligence (AI) is vast, as is the potential for harm.
Diverse, Global and Amortised Counterfactual Explanations for Uncertainty Estimates
To interpret uncertainty estimates from differentiable probabilistic models, recent work has proposed generating a single Counterfactual Latent Uncertainty Explanation (CLUE) for a given data point where the model is uncertain, identifying a single, on-manifold change to the input such that the model becomes more certain in its prediction.
Do Invariances in Deep Neural Networks Align with Human Perception?
One necessary criterion for a network's invariances to align with human perception is for its IRIs look 'similar' to humans.
Towards Principled Disentanglement for Domain Generalization
To tackle this challenge, we first formalize the OOD generalization problem as constrained optimization, called Disentanglement-constrained Domain Generalization (DDG).
PolyViT: Co-training Vision Transformers on Images, Videos and Audio
Can we train a single transformer model capable of processing multiple modalities and datasets, whilst sharing almost all of its learnable parameters?
Provable Defense Against Clustering Attacks on 3D Point Clouds
Lately, the literature on adversarial robustness spans from images to other domains such as point clouds.
Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence
Artificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses.
Iterative Teaching by Label Synthesis
In this paper, we consider the problem of iterative machine teaching, where a teacher provides examples sequentially based on the current iterative learner.
Hybrid Random Features
We propose a new class of random feature methods for linearizing softmax and Gaussian kernels called hybrid random features (HRFs) that automatically adapt the quality of kernel estimation to provide most accurate approximation in the defined regions of interest.
SphereFace Revived: Unifying Hyperspherical Face Recognition
As one of the earliest works in hyperspherical face recognition, SphereFace explicitly proposed to learn face embeddings with large inter-class angular margin.
SphereFace2: Binary Classification is All You Need for Deep Face Recognition
In this paper, we start by identifying the discrepancy between training and evaluation in the existing multi-class classification framework and then discuss the potential limitations caused by the "competitive" nature of softmax normalization.
From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers
In this paper we provide, to the best of our knowledge, the first comprehensive approach for incorporating various masking mechanisms into Transformers architectures in a scalable way.
DIVINE: Diverse Influential Training Points for Data Visualization and Model Refinement
In this work, we take a step towards finding influential training points that also represent the training data well.
On the Expressive Power of Self-Attention Matrices
Our proof is constructive, enabling us to propose an algorithm for finding adaptive inputs and fixed self-attention parameters in order to approximate a given matrix.
Debiasing a First-order Heuristic for Approximate Bi-level Optimization
Approximate bi-level optimization (ABLO) consists of (outer-level) optimization problems, involving numerical (inner-level) optimization loops.
Do Concept Bottleneck Models Learn as Intended?
Concept bottleneck models map from raw inputs to concepts, and then from concepts to targets.
CrossWalk: Fairness-enhanced Node Representation Learning
The potential for machine learning systems to amplify social inequities and unfairness is receiving increasing popular and academic attention.
Is Disentanglement all you need? Comparing Concept-based & Disentanglement Approaches
Concept-based explanations have emerged as a popular way of extracting human-interpretable representations from deep discriminative models.
δ-CLUE: Diverse Sets of Explanations for Uncertainty Estimates
To interpret uncertainty estimates from differentiable probabilistic models, recent work has proposed generating Counterfactual Latent Uncertainty Explanations (CLUEs).
Learning with Hyperspherical Uniformity
Due to the over-parameterization nature, neural networks are a powerful tool for nonlinear function approximation.
Unlocking Pixels for Reinforcement Learning via Implicit Attention
There has recently been significant interest in training reinforcement learning (RL) agents in vision-based environments.
Sub-Linear Memory: How to Make Performers SLiM
Recent works proposed various linear self-attention mechanisms, scaling only as $O(L)$ for serial computation.
Improving Interpretability in Medical Imaging Diagnosis using Adversarial Training
We investigate the influence of adversarial training on the interpretability of convolutional neural networks (CNNs), specifically applied to diagnosing skin cancer.
Ode to an ODE
We present a new paradigm for Neural ODE algorithms, called ODEtoODE, where time-dependent parameters of the main flow evolve according to a matrix flow on the orthogonal group O(d).
Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty
Explainability attempts to provide reasons for a machine learning model's behavior to stakeholders.
Now You See Me (CME): Concept-based Model Extraction
Deep Neural Networks (DNNs) have achieved remarkable performance on a range of tasks.
On the Fairness of Causal Algorithmic Recourse
Algorithmic fairness is typically studied from the perspective of predictions.
Rethinking Attention with Performers
We introduce Performers, Transformer architectures which can estimate regular (softmax) full-rank-attention Transformers with provable accuracy, but using only linear (as opposed to quadratic) space and time complexity, without relying on any priors such as sparsity or low-rankness.
Ranked #7 on
Offline RL
on D4RL
Machine Learning Explainability for External Stakeholders
As machine learning is increasingly deployed in high-stakes contexts affecting people's livelihoods, there have been growing calls to open the black box and to make machine learning algorithms more explainable.
Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch
We study the inverse reinforcement learning (IRL) problem under a transition dynamics mismatch between the expert and the learner.
An Ode to an ODE
We present a new paradigm for Neural ODE algorithms, called ODEtoODE, where time-dependent parameters of the main flow evolve according to a matrix flow on the orthogonal group O(d).
Getting a CLUE: A Method for Explaining Uncertainty Estimates
Both uncertainty estimation and interpretability are important factors for trustworthy machine learning systems.
UFO-BLO: Unbiased First-Order Bilevel Optimization
Bilevel optimization (BLO) is a popular approach with many applications including hyperparameter optimization, neural architecture search, adversarial robustness and model-agnostic meta-learning.
Masked Language Modeling for Proteins via Linearly Scalable Long-Context Transformers
In response, solutions that exploit the structure and sparsity of the learned attention matrix have blossomed.
Adversarial Graph Embeddings for Fair Influence Maximization over Social Networks
We believe we are the first to use embeddings for the task of fair influence maximization.
Time Dependence in Non-Autonomous Neural ODEs
Neural Ordinary Differential Equations (ODEs) are elegant reinterpretations of deep networks where continuous time can replace the discrete notion of depth, ODE solvers perform forward propagation, and the adjoint method enables efficient, constant memory backpropagation.
Dimensions of Diversity in Human Perceptions of Algorithmic Fairness
A growing number of oversight boards and regulatory bodies seek to monitor and govern algorithms that make decisions about people's lives.
Evaluating and Aggregating Feature-based Model Explanations
A feature-based model explanation denotes how much each input feature contributes to a model's output for a given data point.
CWY Parametrization: a Solution for Parallelized Optimization of Orthogonal and Stiefel Matrices
We introduce an efficient approach for optimization over orthogonal groups on highly parallel computation units such as GPUs or TPUs.
Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims
With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure responsible AI development.
Computers and Society
Orthogonal Over-Parameterized Training
The inductive bias of a neural network is largely determined by the architecture and the training algorithm.
Stochastic Flows and Geometric Optimization on the Orthogonal Group
We present a new class of stochastic, geometrically-driven optimization algorithms on the orthogonal group $O(d)$ and naturally reductive homogeneous manifolds obtained from the action of the rotation group $SO(d)$.
DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learning
We introduce a framework for dynamic adversarial discovery of information (DADI), motivated by a scenario where information (a feature set) is used by third parties with unknown objectives.
An Empirical Study on Learning Fairness Metrics for COMPAS Data with Human Supervision
We do not provide a way to directly learn a similarity metric satisfying the individual fairness, but to provide an empirical study on how to derive the similarity metric from human supervisors, then future work can use this as a tool to understand human supervision.
Optimal experimental design via Bayesian optimization: active causal structure learning for Gaussian process networks
We study the problem of causal discovery through targeted interventions.
Exploring Properties of the Deep Image Prior
Finally, we examine the adversarial invariancy of the early DIP outputs, and hypothesize that these outputs may remove non-robust image features.
Explainable Machine Learning in Deployment
Yet there is little understanding of how organizations use these methods in practice.
The Sensitivity of Counterfactual Fairness to Unmeasured Confounding
We demonstrate our new sensitivity analysis tools in real-world fairness scenarios to assess the bias arising from confounding.
Leader Stochastic Gradient Descent for Distributed Training of Deep Learning Models: Extension
Finally, we implement an asynchronous version of our algorithm and extend it to the multi-leader setting, where we form groups of workers, each represented by its own local leader (the best performer in a group), and update each worker with a corrective direction comprised of two attractive forces: one to the local, and one to the global leader (the best performer among all workers).
Orthogonal Estimation of Wasserstein Distances
Wasserstein distances are increasingly used in a wide variety of applications in machine learning.
Self-Guided Belief Propagation -- A Homotopy Continuation Method
Belief propagation (BP) is a popular method for performing probabilistic inference on graphical models.
Geometrically Coupled Monte Carlo Sampling
Monte Carlo sampling in high-dimensional, low-sample settings is important in many machine learning tasks.
Proceedings of the 2018 ICML Workshop on Human Interpretability in Machine Learning (WHI 2018)
This is the Proceedings of the 2018 ICML Workshop on Human Interpretability in Machine Learning (WHI 2018), which was held in Stockholm, Sweden, July 14, 2018.
A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices
Further, our work reveals overlooked tradeoffs between different fairness notions: using our proposed measures, the overall individual-level unfairness of an algorithm can be decomposed into a between-group and a within-group component.
Discovering Interpretable Representations for Both Deep Generative and Discriminative Models
We use a generative model which takes as input the representation in an existing (generative or discriminative) model, weakly supervised by limited side information.
Blind Justice: Fairness with Encrypted Sensitive Attributes
Recent work has explored how to train machine learning models which do not discriminate against any subgroup of the population as determined by sensitive attributes such as gender or race.
Structured Evolution with Compact Architectures for Scalable Policy Optimization
We present a new method of blackbox optimization via gradient approximation with the use of structured random orthogonal matrices, providing more accurate estimators than baselines and with provable theoretical guarantees.
Bucket Renormalization for Approximate Inference
Probabilistic graphical models are a key tool in machine learning applications.
Human Perceptions of Fairness in Algorithmic Decision Making: A Case Study of Criminal Risk Prediction
As algorithms are increasingly used to make important decisions that affect human lives, ranging from social benefit assignment to predicting risk of criminal recidivism, concerns have been raised about the fairness of algorithmic decision making.
Gauged Mini-Bucket Elimination for Approximate Inference
Recently, so-called gauge transformations were used to improve variational lower bounds on $Z$.
Uprooting and Rerooting Higher-Order Graphical Models
The idea of uprooting and rerooting graphical models was introduced specifically for binary pairwise models by Weller (2016) as a way to transform a model to any of a whole equivalence class of related models, such that inference on any one model yields inference results for all others.
Accountability of AI Under the Law: The Role of Explanation
The ubiquity of systems using artificial intelligence or "AI" has brought increasing attention to how those systems should be regulated.
Proceedings of the 2017 ICML Workshop on Human Interpretability in Machine Learning (WHI 2017)
This is the Proceedings of the 2017 ICML Workshop on Human Interpretability in Machine Learning (WHI 2017), which was held in Sydney, Australia, August 10, 2017.
Challenges for Transparency
Transparency is often deemed critical to enable effective real-world deployment of intelligent systems.
Computers and Society
From Parity to Preference-based Notions of Fairness in Classification
The adoption of automated, data-driven decision making in an ever expanding range of applications has raised concerns about its potential unfairness towards certain social groups.
On Fairness, Diversity and Randomness in Algorithmic Decision Making
Consider a binary decision making process where a single machine learning classifier replaces a multitude of humans.
Lost Relatives of the Gumbel Trick
We show how a subfamily of our new methods adapts to this setting, proving new upper and lower bounds on the log partition function and deriving a family of sequential samplers for the Gibbs distribution.
The Unreasonable Effectiveness of Structured Random Orthogonal Embeddings
We examine a class of embeddings based on structured random matrices with orthogonal rows which can be applied in many machine learning applications including dimensionality reduction and kernel approximation.
Train and Test Tightness of LP Relaxations in Structured Prediction
Structured prediction is used in areas such as computer vision and natural language processing to predict structured outputs such as segmentations or parse trees.
Clamping Improves TRW and Mean Field Approximations
We examine the effect of clamping variables for approximate inference in undirected graphical models with pairwise relationships and discrete variables.
Clamping Variables and Approximate Inference
It was recently proved using graph covers (Ruozzi, 2012) that the Bethe partition function is upper bounded by the true partition function for a binary pairwise model that is attractive.
Approximating the Bethe partition function
When belief propagation (BP) converges, it does so to a stationary point of the Bethe free energy $F$, and is often strikingly accurate.
On MAP Inference by MWSS on Perfect Graphs
Finding the most likely (MAP) configuration of a Markov random field (MRF) is NP-hard in general.
