Skip to yearly menu bar Skip to main content


Show Detail Timezone:
America/Los_Angeles
 
Filter Rooms:  

SUN 10 DEC
7 a.m.
(ends 4:00 PM)
noon
Expo Demonstration:
(duration 1.5 hr)
Expo Talk Panel:
(ends 12:50 PM)
Expo Talk Panel:
(ends 12:50 PM)
Expo Workshop:
(ends 2:00 PM)
Expo Workshop:
(ends 2:00 PM)
Expo Workshop:
(ends 2:00 PM)
Expo Workshop:
(ends 2:00 PM)
1 p.m.
Break:
(ends 1:30 PM)
2:30 p.m.
Break:
(ends 3:00 PM)

MON 11 DEC
5:30 a.m.
(ends 4:00 PM)
6 a.m.
Workshop:
(ends 1:30 PM)
6:15 a.m.
Workshop:
(ends 1:30 PM)
Workshop:
(ends 1:30 PM)
Workshop:
(ends 1:30 PM)
6:30 a.m.
Workshop:
(ends 1:30 PM)
Workshop:
(ends 1:30 PM)
7 a.m.
Workshop:
(ends 1:15 PM)
7:15 a.m.
Break:
(ends 8:05 AM)
7:30 a.m.
Workshop:
(ends 1:30 PM)
9 a.m.
Workshop:
(ends 12:00 PM)
10 a.m.
Education Outreach:
(ends 11:40 AM)
10:15 a.m.
Lunch:
(ends 11:45 AM)
1:30 p.m.
Affinity Poster Session:
(ends 2:30 PM)
2 p.m.
Break:
(ends 2:45 PM)
3 p.m.
Opening Remarks:
(ends 3:25 PM)
3:25 p.m.
Invited Talk:
Björn Ommer
(ends 4:15 PM)
4:15 p.m.
Reception:
(ends 6:30 PM)
4:30 p.m.
Chairs: Jean Oh · Isabelle Guyon
s 4:30-6:00
[4:30] Voice Scroll
[4:30] The WHOOPS! Gallery: An Intersection of AI, Creativity, and the Unusual
[4:30] Kiss/Crash
[4:30] Emergent Rhythm — Real-time AI Generative DJ Set
[4:30] Fusion: Landscape and Beyond
[4:30] Entanglement
(ends 6:00 PM)

TUE 12 DEC
5:30 a.m.
(ends 4:00 PM)
6:30 a.m.
Invited Talk:
Lora Aroyo
(ends 7:20 AM)
7:15 a.m.
Break:
(ends 8:05 AM)
8 a.m.
Orals 8:00-8:45
[8:00] Ordering-based Conditions for Global Convergence of Policy Gradient Methods
[8:15] When Demonstrations meet Generative World Models: A Maximum Likelihood Framework for Offline Inverse Reinforcement Learning
[8:30] Online RL in Linearly $q^\pi$-Realizable MDPs Is as Easy as in Linear MDPs If You Learn What to Ignore
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] LeanDojo: Theorem Proving with Retrieval-Augmented Language Models
[8:15] OpenAssistant Conversations - Democratizing Large Language Model Alignment
[8:30] DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] How to Turn Your Knowledge Graph Embeddings into Generative Models
[8:15] Exact Bayesian Inference on Discrete Models via Probability Generating Functions: A Probabilistic Programming Approach
[8:30] Characteristic Circuits
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] Sharpness Minimization Algorithms Do Not Only Minimize Sharpness To Achieve Better Generalization
[8:15] Abide by the law and follow the flow: conservation laws for gradient flows
[8:30] A U-turn on Double Descent: Rethinking Parameter Counting in Statistical Learning
(ends 8:45 AM)
8:45 a.m.
Posters 8:45-10:45
(ends 10:45 AM)
10:45 a.m.
Lunch:
(ends 12:15 PM)
1:05 p.m.
Chairs: Jean Oh · Isabelle Guyon
s 1:05-1:40
[1:05] Kiss/Crash
[1:05] Blabrecs: An AI-Based Game of Nonsense Word Creation
[1:05] Androgynous and Mixed Race Human Face
[1:05] Voice Scroll
[1:05] The WHOOPS! Gallery: An Intersection of AI, Creativity, and the Unusual
[1:05] Visions of Resilience: Augmented Diversity
[1:05] salad bowl
[1:05] AI Applications to Illustrate Native American Arts: Birdsongs: Using Transfer Learning to Augment Image Generation Models
[1:05] Visualising AI
[1:05] Creating playful comics together with AI
(ends 1:40 PM)
1:20 p.m.
Break:
(ends 1:45 PM)
1:40 p.m.
Orals 1:40-2:40
[1:40] Monarch Mixer: A Simple Sub-Quadratic GEMM-Based Architecture
[1:55] QLoRA: Efficient Finetuning of Quantized LLMs
[2:10] Scaling Data-Constrained Language Models
[2:25] Bridging Discrete and Backpropagation: Straight-Through and Beyond
(ends 2:40 PM)
Orals 1:40-2:40
[1:40] Rotating Features for Object Discovery
[1:55] Linguistic Binding in Diffusion Models: Enhancing Attribute Correspondence through Attention Map Alignment
[2:10] Additive Decoders for Latent Variables Identification and Cartesian-Product Extrapolation
[2:25] Emergence of Shape Bias in Convolutional Neural Networks through Activation Sparsity
(ends 2:40 PM)
Orals 1:40-2:40
[1:40] Learning Linear Causal Representations from Interventions under General Nonlinear Mixing
[1:55] A Measure-Theoretic Axiomatisation of Causality
[2:10] Conformal Meta-learners for Predictive Inference of Individual Treatment Effects
[2:25] Causal normalizing flows: from theory to practice
(ends 2:40 PM)
Orals 1:40-2:40
[1:40] Nearly Tight Bounds For Differentially Private Multiway Cut
[1:55] Privacy Auditing with One (1) Training Run
[2:10] Private Everlasting Prediction
[2:25] User-Level Differential Privacy With Few Examples Per User
(ends 2:40 PM)
2:45 p.m.
Break:
(ends 3:30 PM)
3:15 p.m.
Posters 3:15-5:15
(ends 5:15 PM)

WED 13 DEC
5:30 a.m.
(ends 4:00 PM)
6:30 a.m.
7:15 a.m.
Break:
(ends 8:05 AM)
8 a.m.
Orals 8:00-8:45
[8:00] Brain Diffusion for Visual Exploration: Cortical Discovery using Large Scale Generative Models
[8:15] Cinematic Mindscapes: High-quality Video Reconstruction from Brain Activity
[8:30] Human-like Few-Shot Learning via Bayesian Reasoning over Natural Language
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings
[8:15] Toolformer: Language Models Can Teach Themselves to Use Tools
[8:30] Learning Transformer Programs
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] Understanding Diffusion Objectives as the ELBO with Simple Data Augmentation
[8:15] Entropic Neural Optimal Transport via Diffusion Processes
[8:30] DiffuseBot: Breeding Soft Robots With Physics-Augmented Generative Diffusion Models
(ends 8:45 AM)
8:45 a.m.
Posters 8:45-10:45
(ends 10:45 AM)
Chairs: Jean Oh · Isabelle Guyon
s 8:45-12:15
[8:45] Kiss/Crash
[8:45] Blabrecs: An AI-Based Game of Nonsense Word Creation
[8:45] Androgynous and Mixed Race Human Face
[8:45] Voice Scroll
[8:45] The WHOOPS! Gallery: An Intersection of AI, Creativity, and the Unusual
[8:45] Visions of Resilience: Augmented Diversity
[8:45] salad bowl
[8:45] Creating playful comics together with AI
[8:45] AI Applications to Illustrate Native American Arts: Birdsongs: Using Transfer Learning to Augment Image Generation Models
[8:45] Visualising AI
(ends 12:15 PM)
10:45 a.m.
Lunch:
(ends 12:15 PM)
12:15 p.m.
Invited Talk:
Alexander Rush · Aakanksha Chowdhery · Angela Fan · Percy Liang · Jie Tang
(ends 1:15 PM)
1:15 p.m.
Break:
(ends 1:45 PM)
1:30 p.m.
Orals 1:30-2:30
[1:30] A Single-Loop Accelerated Extra-Gradient Difference Algorithm with Improved Complexity Bounds for Constrained Minimax Optimization
[1:45] Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Models
[2:00] Generalizing Nonlinear ICA Beyond Structural Sparsity
[2:15] Fine-Tuning Language Models with Just Forward Passes
(ends 2:30 PM)
Orals 1:30-2:30
[1:30] Mesogeos: A multi-purpose dataset for data-driven wildfire modeling in the Mediterranean
[1:45] ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation
[2:00] Quilt-1M: One Million Image-Text Pairs for Histopathology
[2:15] BEDD: The MineRL BASALT Evaluation and Demonstrations Dataset for Training and Benchmarking Agents that Solve Fuzzy Tasks
(ends 2:30 PM)
Orals 1:30-2:30
[1:30] Why think step by step? Reasoning emerges from the locality of experience
[1:45] Tree of Thoughts: Deliberate Problem Solving with Large Language Models
[2:00] Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection
[2:15] Towards Revealing the Mystery behind Chain of Thought: A Theoretical Perspective
(ends 2:30 PM)
2:15 p.m.
Break:
(ends 3:00 PM)
3 p.m.
Posters 3:00-5:00
(ends 5:00 PM)
5 p.m.

THU 14 DEC
5:30 a.m.
(ends 4:00 PM)
6:30 a.m.
Invited Talk:
Christopher Ré
(ends 7:20 AM)
7:15 a.m.
Break:
(ends 8:05 AM)
8 a.m.
Orals 8:00-8:45
[8:00] Going beyond persistent homology using persistent homology
[8:15] Clifford Group Equivariant Neural Networks
[8:30] Evaluating Post-hoc Explanations for Graph Neural Networks via Robustness Analysis
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] Students Parrot Their Teachers: Membership Inference on Model Distillation
[8:15] Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face Recognition
[8:30] Ethical Considerations for Responsible Data Curation
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] Sampling from Gaussian Process Posteriors using Stochastic Gradient Descent
[8:15] A Rigorous Link between Deep Ensembles and (Variational) Bayesian Methods
[8:30] Optimizing Solution-Samplers for Combinatorial Problems: The Landscape of Policy-Gradient Method
(ends 8:45 AM)
Orals 8:00-8:45
[8:00] Visual Instruction Tuning
[8:15] EgoEnv: Human-centric environment representations from egocentric video
[8:30] DataComp: In search of the next generation of multimodal datasets
(ends 8:45 AM)
8:45 a.m.
Posters 8:45-10:45
(ends 10:45 AM)
Chairs: Jean Oh · Isabelle Guyon
s 8:45-12:15
[8:45] Archi Guesser - An AI Art Architecture Educational Game
[8:45] Collaborative Synthscapes from Words
[8:45] Creating Playful Comics Together with AI
[8:45] Q's Views #1-#3 (+1)
[8:45] Resonator: An AI-assisted Musical Experience for Human Connection
[8:45] salad bowl
[8:45] NeuroView: Generative Visualization of the Diversity of Brain Responses to Jazz
[8:45] Fusion: Landscape and Beyond
(ends 12:15 PM)
10:45 a.m.
Lunch:
(ends 12:15 PM)
12:15 p.m.
Invited Talk:
Susan Murphy
(ends 1:05 PM)
1 p.m.
Break:
(ends 1:25 PM)
1:20 p.m.
Orals 1:20-2:20
[1:20] Are Emergent Abilities of Large Language Models a Mirage?
[1:35] Task Arithmetic in the Tangent Space: Improved Editing of Pre-Trained Models
[1:50] The Clock and the Pizza: Two Stories in Mechanistic Explanation of Neural Networks
[2:05] Jailbroken: How Does LLM Safety Training Fail?
(ends 2:20 PM)
Orals 1:20-2:20
[1:20] When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
[1:35] Bridging RL Theory and Practice with the Effective Horizon
[1:50] Direct Preference Optimization: Your Language Model is Secretly a Reward Model
[2:05] MetaBox: A Benchmark Platform for Meta-Black-Box Optimization with Reinforcement Learning
(ends 2:20 PM)
Orals 1:20-2:20
[1:20] Siamese Masked Autoencoders
[1:35] Image Captioners Are Scalable Vision Learners Too
[1:50] The Surprising Effectiveness of Diffusion Models for Optical Flow and Monocular Depth Estimation
[2:05] Spatial-frequency channels, shape bias, and adversarial robustness
(ends 2:20 PM)
Orals 1:20-2:20
[1:20] Optimal Learners for Realizable Regression: PAC Learning and Online Learning
[1:35] Random Cuts are Optimal for Explainable k-Medians
[1:50] Tester-Learners for Halfspaces: Universal Algorithms
[2:05] Improved Algorithms for Stochastic Linear Bandits Using Tail Bounds for Martingale Mixtures
(ends 2:20 PM)
2:20 p.m.
Break:
(ends 3:15 PM)
3 p.m.
Posters 3:00-5:00
(ends 5:00 PM)
5 p.m.
Chairs: Jean Oh · Isabelle Guyon
s 5:00-6:45
[5:00] Resonator: An AI-assisted Musical Experience for Human Connection
[5:00] Fusion: Landscape and Beyond
[5:00] Emergent Rhythm — Real-time AI Generative DJ Set
[5:00] The WHOOPS! Gallery: An Intersection of AI, Creativity, and the Unusual
[5:00] Entanglement
[5:00] Kiss/Crash
[6:00] Collaborative Synthscapes from Words
(ends 7:00 PM)
Town Hall:
(ends 6:00 PM)
6 p.m.
Reception:
(ends 7:00 PM)

FRI 15 DEC
5:30 a.m.
(ends 2:00 PM)
6:35 a.m.
Competition:
(ends 10:00 AM)
8 a.m.
Competition:
(ends 12:00 PM)
Break:
(ends 8:30 AM)
10 a.m.
Break:
(ends 11:30 AM)
1 p.m.
Break:
(ends 1:30 PM)
3:30 p.m.

SAT 16 DEC
5:30 a.m.
(ends 10:00 AM)
6:20 a.m.
Workshop:
(ends 3:30 PM)
8 a.m.
Break:
(ends 8:30 AM)
10 a.m.
Break:
(ends 11:30 AM)
1 p.m.
Break:
(ends 1:30 PM)

TUE 19 DEC
7 a.m.
Affinity Poster Session:
(ends 10:00 AM)
7 p.m.
Affinity Poster Session:
(ends 9:59 PM)