Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work emphasizes the limitations of RLHF and highlights the importance of a multi-faceted approach to the development of safer AI systems.
We propose the Thinker algorithm, a novel approach that enables reinforcement learning agents to autonomously interact with and utilize a learned world model. The Thinker algorithm wraps the environment with a world model and introduces new actions designed for interacting with the world model. These model-interaction actions enable agents to perform planning by proposing alternative plans to the world model before selecting a final action to execute in the environment. This approach eliminates the need for hand-crafted planning algorithms by enabling the agent to learn how to plan autonomously and allows for easy interpretation of the agent's plan with visualization. We demonstrate the algorithm's effectiveness through experimental results in the game of Sokoban and the Atari 2600 benchmark, where the Thinker algorithm achieves state-of-the-art performance and competitive results, respectively. Visualizations of agents trained with the Thinker algorithm demonstrate that they have learned to plan effectively with the world model to select better actions. The algorithm's generality opens a new research direction on how a world model can be used in reinforcement learning and how planning can be seamlessly integrated into an agent's decision-making process.
Visual object recognition systems need to generalize from a set of 2D training views to novel views. The question of how the human visual system can generalize to novel views has been studied and modeled in psychology, computer vision, and neuroscience. Modern deep learning architectures for object recognition generalize well to novel views, but the mechanisms are not well understood. In this paper, we characterize the ability of common deep learning architectures to generalize to novel views. We formulate this as a supervised classification task where labels correspond to unique 3D objects and examples correspond to 2D views of the objects at different 3D orientations. We consider three common models of generalization to novel views: (i) full 3D generalization, (ii) pure 2D matching, and (iii) matching based on a linear combination of views. We find that deep models generalize well to novel views, but they do so in a way that differs from all these existing models. Extrapolation to views beyond the range covered by views in the training set is limited, and extrapolation to novel rotation axes is even more limited, implying that the networks do not infer full 3D structure, nor use linear interpolation. Yet, generalization is far superior to pure 2D matching. These findings help with designing datasets with 2D views required to achieve 3D generalization. Code to reproduce our experiments is publicly available: https://github.com/shoaibahmed/investigating_3d_generalization.git
A principled understanding of generalization in deep learning may require unifying disparate observations under a single conceptual framework. Previous work has studied \emph{grokking}, a training dynamic in which a sustained period of near-perfect training performance and near-chance test performance is eventually followed by generalization, as well as the superficially similar \emph{double descent}. These topics have so far been studied in isolation. We hypothesize that grokking and double descent can be understood as instances of the same learning dynamics within a framework of pattern learning speeds. We propose that this framework also applies when varying model capacity instead of optimization steps, and provide the first demonstration of model-wise grokking.
Current state-of-the-art deep networks are all powered by backpropagation. In this paper, we explore alternatives to full backpropagation in the form of blockwise learning rules, leveraging the latest developments in self-supervised learning. We show that a blockwise pretraining procedure consisting of training independently the 4 main blocks of layers of a ResNet-50 with Barlow Twins' loss function at each block performs almost as well as end-to-end backpropagation on ImageNet: a linear probe trained on top of our blockwise pretrained model obtains a top-1 classification accuracy of 70.48%, only 1.1% below the accuracy of an end-to-end pretrained network (71.57% accuracy). We perform extensive experiments to understand the impact of different components within our method and explore a variety of adaptations of self-supervised learning to the blockwise paradigm, building an exhaustive understanding of the critical avenues for scaling local learning rules to large networks, with implications ranging from hardware design to neuroscience.
Reward functions are notoriously difficult to specify, especially for tasks with complex goals. Reward learning approaches attempt to infer reward functions from human feedback and preferences. Prior works on reward learning have mainly focused on the performance of policies trained alongside the reward function. This practice, however, may fail to detect learned rewards that are not capable of training new policies from scratch and thus do not capture the intended behavior. Our work focuses on demonstrating and studying the causes of these relearning failures in the domain of preference-based reward learning. We demonstrate with experiments in tabular and continuous control environments that the severity of relearning failures can be sensitive to changes in reward model design and the trajectory dataset composition. Based on our findings, we emphasize the need for more retraining-based evaluations in the literature.
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. We consider multi-demonstrator offline RL, a middle ground where we know which demonstrators generated each dataset, but make no assumptions about the underlying policies of the demonstrators. This is the most natural setting when collecting data from multiple human operators, yet remains unexplored. Since different demonstrators induce different data distributions, we show that this can be naturally framed as a domain generalization problem, with each demonstrator corresponding to a different domain. Specifically, we propose Domain-Invariant Model-based Offline RL (DIMORL), where we apply Risk Extrapolation (REx) (Krueger et al., 2020) to the process of learning dynamics and rewards models. Our results show that models trained with REx exhibit improved domain generalization performance when compared with the natural baseline of pooling all demonstrators' data. We observe that the resulting models frequently enable the learning of superior policies in the offline model-based RL setting, can improve the stability of the policy learning process, and potentially enable increased exploration.
Neural networks are known to be biased towards learning mechanisms that help identify $spurious\, attributes$, yielding features that do not generalize well under distribution shifts. To understand and address this limitation, we study the geometry of neural network loss landscapes through the lens of $mode\, connectivity$, the observation that minimizers of neural networks are connected via simple paths of low loss. Our work addresses two questions: (i) do minimizers that encode dissimilar mechanisms connect via simple paths of low loss? (ii) can fine-tuning a pretrained model help switch between such minimizers? We define a notion of $\textit{mechanistic similarity}$ and demonstrate that lack of linear connectivity between two minimizers implies the corresponding models use dissimilar mechanisms for making their predictions. This property helps us demonstrate that na$\"{i}$ve fine-tuning can fail to eliminate a model's reliance on spurious attributes. We thus propose a method for altering a model's mechanisms, named $connectivity$-$based$ $fine$-$tuning$, and validate its usefulness by inducing models invariant to spurious attributes.
We present a smoothly broken power law functional form that accurately models and extrapolates the scaling behaviors of deep neural networks (i.e. how the evaluation metric of interest varies as the amount of compute used for training, number of model parameters, training dataset size, or upstream performance varies) for each task within a large and diverse set of upstream and downstream tasks, in zero-shot, prompted, and fine-tuned settings. This set includes large-scale vision and unsupervised language tasks, diffusion generative modeling of images, arithmetic, and reinforcement learning. When compared to other functional forms for neural scaling behavior, this functional form yields extrapolations of scaling behavior that are considerably more accurate on this set. 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. Lastly, we use this functional form to glean insights about the limit of the predictability of scaling behavior. Code is available at https://github.com/ethancaballero/broken_neural_scaling_laws