Search Results for author:
Found 31 papers, 15 papers with code
Investigating the Nature of 3D Generalization in Deep Neural Networks
Modern deep learning architectures for object recognition generalize well to novel views, but the mechanisms are not well understood.
Unifying Grokking and Double Descent
A principled understanding of generalization in deep learning may require unifying disparate observations under a single conceptual framework.
Blockwise Self-Supervised Learning at Scale
Current state-of-the-art deep networks are all powered by backpropagation.
On The Fragility of Learned Reward Functions
Our work focuses on demonstrating and studying the causes of these relearning failures in the domain of preference-based reward learning.
Domain Generalization for Robust Model-Based Offline Reinforcement Learning
Existing offline reinforcement learning (RL) algorithms typically assume that training data is either: 1) generated by a known policy, or 2) of entirely unknown origin.
Mechanistic Mode Connectivity
We study neural network loss landscapes through the lens of mode connectivity, the observation that minimizers of neural networks retrieved via training on a dataset are connected via simple paths of low loss.
Broken Neural Scaling Laws
Moreover, this functional form accurately models and extrapolates scaling behavior that other functional forms are incapable of expressing such as the non-monotonic transitions present in the scaling behavior of phenomena such as double descent and the delayed, sharp inflection points present in the scaling behavior of tasks such as arithmetic.
Towards Out-of-Distribution Adversarial Robustness
Compared to existing methods, we obtain similar or superior worst-case adversarial robustness on attacks seen during training.
Defining and Characterizing Reward Hacking
We provide the first formal definition of reward hacking, a phenomenon where optimizing an imperfect proxy reward function, $\mathcal{\tilde{R}}$, leads to poor performance according to the true reward function, $\mathcal{R}$.
Metadata Archaeology: Unearthing Data Subsets by Leveraging Training Dynamics
Modern machine learning research relies on relatively few carefully curated datasets.
Multi-Domain Balanced Sampling Improves Out-of-Distribution Generalization of Chest X-ray Pathology Prediction Models
Learning models that generalize under different distribution shifts in medical imaging has been a long-standing research challenge.
Filling gaps in trustworthy development of AI
The range of application of artificial intelligence (AI) is vast, as is the potential for harm.
Revealing the Incentive to Cause Distributional Shift
We use these unit tests to demonstrate that changes to the learning algorithm (e. g. introducing meta-learning) can cause previously hidden incentives to be revealed, resulting in qualitatively different behaviour despite no change in performance metric.
Goal Misgeneralization in Deep Reinforcement Learning
We study goal misgeneralization, a type of out-of-distribution generalization failure in reinforcement learning (RL).
Active Reinforcement Learning: Observing Rewards at a Cost
Active reinforcement learning (ARL) is a variant on reinforcement learning where the agent does not observe the reward unless it chooses to pay a query cost c > 0.
Hidden Incentives for Auto-Induced Distributional Shift
We introduce the term auto-induced distributional shift (ADS) to describe the phenomenon of an algorithm causing a change in the distribution of its own inputs.
AI Research Considerations for Human Existential Safety (ARCHES)
Framed in positive terms, this report examines how technical AI research might be steered in a manner that is more attentive to humanity's long-term prospects for survival as a species.
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
Out-of-Distribution Generalization via Risk Extrapolation (REx)
Distributional shift is one of the major obstacles when transferring machine learning prediction systems from the lab to the real world.
Scalable agent alignment via reward modeling: a research direction
One obstacle to applying reinforcement learning algorithms to real-world problems is the lack of suitable reward functions.
Uncertainty in Multitask Transfer Learning
The result is a rich and meaningful prior capable of few-shot learning on new tasks.
Neural Autoregressive Flows
Normalizing flows and autoregressive models have been successfully combined to produce state-of-the-art results in density estimation, via Masked Autoregressive Flows (MAF), and to accelerate state-of-the-art WaveNet-based speech synthesis to 20x faster than real-time, via Inverse Autoregressive Flows (IAF).
Nested LSTMs
We propose Nested LSTMs (NLSTM), a novel RNN architecture with multiple levels of memory.
Deep Prior
The recent literature on deep learning offers new tools to learn a rich probability distribution over high dimensional data such as images or sounds.
Bayesian Hypernetworks
We study Bayesian hypernetworks: a framework for approximate Bayesian inference in neural networks.
A Closer Look at Memorization in Deep Networks
We examine the role of memorization in deep learning, drawing connections to capacity, generalization, and adversarial robustness.
Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations
We propose zoneout, a novel method for regularizing RNNs.
Regularizing RNNs by Stabilizing Activations
We stabilize the activations of Recurrent Neural Networks (RNNs) by penalizing the squared distance between successive hidden states' norms.
Testing Visual Attention in Dynamic Environments
We investigate attention as the active pursuit of useful information.
NICE: Non-linear Independent Components Estimation
It is based on the idea that a good representation is one in which the data has a distribution that is easy to model.
Ranked #69 on
Image Generation
on CIFAR-10
(bits/dimension metric)
Zero-bias autoencoders and the benefits of co-adapting features
We show that negative biases are a natural result of using a hidden layer whose responsibility is to both represent the input data and act as a selection mechanism that ensures sparsity of the representation.
