[ Logo] Artificial Intelligence and Statistics 2023


Program Schedule

All times are CEST. You can check current CEST time here.

Registration Desk

Registration desk is open on:

Schedule for Day 1: Tue, April 25

Time (CEST) Day 1: Tue, April 25
08:45-09:00 Opening remarks
09:00-10:00 Keynote Talk: Arthur Gretton (UCL Gatsby)
Causal Effect Estimation with Context and Confounders A fundamental causal modelling task is to predict the effect of an intervention (or treatment) on an outcome, given context/covariates. Examples include predicting the effect of a medical treatment on patient health given patient symptoms and demographic information, or predicting the effect of ticket pricing on airline sales given seasonal fluctuations in demand. The problem becomes especially challenging when the treatment and context are complex (for instance, “treatment” might be a web ad design or a radiotherapy plan), and when only observational data is available (i.e., we have access to historical data, but cannot intervene ourselves). The challenge is greater still when the covariates are not observed, and constitute a hidden source of confounding. I will give an overview of some practical tools and methods for estimating causal effects of complex, high dimensional treatments from observational data. The approach is based on conditional feature means, which represent conditional expectations of relevant model features. These features can be deep neural nets (adaptive, finite dimensional, learned from data), or kernel features (fixed, infinite dimensional, enforcing smoothness). When hidden confounding is present, a neural net implementation of instrumental variable regression can be used to correct for this confounding. The methods will be applied to modelling employment outcomes for the US Job Corps program for Disadvantaged Youth, and in policy evaluation for reinforcement learning.
10:00-10:30 Coffee break
Oral Session 1 | Optimal Transport, Information Theory
  • The Schrödinger Bridge between Gaussian Measures has a Closed Form
  • Rethinking Initialization of the Sinkhorn Algorithm
  • Using Sliced Mutual Information to Study Memorization and Generalization in Deep Neural Networks
  • Mode-Seeking Divergences: Theory and Applications to GANs
Affinity Groups Panel | Carving out a niche as a researcher Panelists: Marc Deisenroth, Jessica Schrouff, Santiago Velasco-Forero, Laura Montoya
Moderator: Francisco J. R. Ruiz
12:30-14:00 Lunch break
Oral Session 2 | Trustworthy ML and Statistics
  • Who Should Predict? Exact Algorithms For Learning to Defer to Humans
  • Generalized PTR: User-Friendly Recipes for Data-Adaptive Algorithms with Differential Privacy
  • Origins of Low-Dimensional Adversarial Perturbations
  • Data Banzhaf: A Robust Data Valuation Framework for Machine Learning
15:00-15:30 Coffee break
Oral Session 3 | Representations of Graphs
  • The Power of Recursion in Graph Neural Networks for Counting Substructures
  • Implicit Graphon Neural Representation
  • Implications of sparsity and high triangle density for graph representation learning
  • Fitting low-rank models on egocentrically sampled partial networks
16:30-19:00 Poster session 1
18:00-20:00 LatinX in AI Social (Main Auditorium)

Schedule for Day 2: Wed, April 26th

Time (CEST) Day 2: Wed, April 26th
09:00-10:00 Keynote Talk: Shakir Mohamed (Deepmind)
Elevating our Evaluations: Technical and Sociotechnical Standards of Assessment in Machine Learning Evaluation in Machine Learning does not always get the attention it deserves. I hope to focus our attention for the time of this talk on the questions of systematic evaluation in machine learning and the changes that we should continue to make as we elevate the standard of evaluation across our field. The breadth of application areas we collaborate on in machine learning requires a variety of approaches for evaluation, and we'll explore this variety by considering applications in generative models, social good, healthcare, and environmental science. Grounded in these applications, we will expand the conceptual aperture through which we think about machine learning evaluations, starting from purely technical evaluations (thinking about likelihoods), moving to mixed methods (with proper scoring rules and expert assessments), and then to sociotechnical assessments (considering fairness, impacts, and participation). My core message is that broad and expansive evaluation remains fundamental and an area into which I hope we will drive even greater investments as a community, together.
10:00-10:30 Coffee break
Oral Session 4 | Probabilistic Methods 1
  • Do Bayesian Neural Networks Need To Be Fully Stochastic?
  • Indeterminacy in Generative Models: Characterization and Strong Identifiability
  • Distance-to-Set Priors and Constrained Bayesian Inference
  • Particle algorithms for maximum likelihood training of latent variable models
Oral Session 5 | Probabilistic Methods 2
  • BaCaDI: Bayesian Causal Discovery with Unknown Interventions
  • Multilevel Bayesian Quadrature
  • Discovering Many Diverse Solutions with Bayesian Optimization
  • Inducing Point Allocation for Sparse Gaussian Processes in High-Throughput Bayesian Optimisation
12:30-14:00 Lunch break
14:00-15:00 Test of Time Award: Andreas Damianou and Neil Lawrence
Deep Gaussian Processes (published at AISTATS 2013)
15:00-15:30 Coffee break
Oral Session 6 | Statistical Methods 1
  • Huber-robust confidence sequences
  • Error Estimation for Random Fourier Features
  • A Tale of Sampling and Estimation in Discounted Reinforcement Learning
  • Safe Sequential Testing and Effect Estimation in Stratified Count Data
16:30-19:00 Poster session 2

Schedule for Day 3: Thu, April 27th

Time (CEST) Day 3: Thu, April 27th
08:00-09:00 Mentoring Event | Tips for Scientific Writing by Marc Deisenroth
09:00-10:00 Keynote Talk: Tamara Broderick (MIT)
An Automatic Finite-Sample Robustness Check: Can Dropping a Little Data Change Conclusions? Practitioners will often analyze a data sample with the goal of applying any conclusions to a new population. For instance, if economists conclude microcredit is effective at alleviating poverty based on observed data, policymakers might decide to distribute microcredit in other locations or future years. Typically, the original data is not a perfect random sample from the population where policy is applied -- but researchers might feel comfortable generalizing anyway so long as deviations from random sampling are small, and the corresponding impact on conclusions is small as well. Conversely, researchers might worry if a very small proportion of the data sample was instrumental to the original conclusion. So we propose a method to assess the sensitivity of statistical conclusions to the removal of a very small fraction of the data set. Manually checking all small data subsets is computationally infeasible, so we propose an approximation based on the classical influence function. Our method is automatically computable for common estimators. We provide finite-sample error bounds on approximation performance and a low-cost exact lower bound on sensitivity. We find that sensitivity is driven by a signal-to-noise ratio in the inference problem, does not disappear asymptotically, and is not decided by misspecification. Empirically we find that many data analyses are robust, but the conclusions of several influential economics papers can be changed by removing (much) less than 1% of the data.
10:00-10:30 Coffee break
Oral Session 7 | Supervised Learning
  • Don't be fooled: label leakage in explanation methods and the importance of their quantitative evaluation
  • Fix-A-Step: Semi-supervised Learning From Uncurated Unlabeled Data
  • Blessing of Class Diversity in Pre-training
  • Federated Learning under Distributed Concept Drift
Oral Session 8 | Statistical Methods 2
  • Scalable Bicriteria Algorithms for Non-Monotone Submodular Cover
  • Noisy Low-rank Matrix Optimization: Geometry of Local Minima and Convergence Rate
  • An Efficient and Continuous Voronoi Density Estimator
  • Hedging against Complexity: Distributionally Robust Optimization with Parametric Approximation
12:30-14:00 Lunch break
14:00-16:30 Poster session 3
This site last compiled Thu, 25 May 2023 21:48:13 -0500
Github Account Copyright © AISTATS 2023. All rights reserved.