Posts in category: Berkeley

<!– Figure title. Figure caption. This image is centered and set to 50%
page width. –>

A longstanding goal of the field of robot learning has been to create generalist agents that can perform tasks for humans. Natural language has the potential to be an easy-to-use interface for humans to specify arbitrary tasks, but it is difficult to train robots to follow language instructions. Approaches like language-conditioned behavioral cloning (LCBC) train policies to directly imitate expert actions conditioned on language, but require humans to annotate all training trajectories and generalize poorly across scenes and behaviors. Meanwhile, recent goal-conditioned approaches perform much better at general manipulation tasks, but do not enable easy task specification for human operators. How can we reconcile the ease of specifying tasks through LCBC-like approaches with the performance improvements of goal-conditioned learning?

<!– Figure title. Figure caption. This image is centered and set to 50%
page width. –>

A longstanding goal of the field of robot learning has been to create generalist agents that can perform tasks for humans. Natural language has the potential to be an easy-to-use interface for humans to specify arbitrary tasks, but it is difficult to train robots to follow language instructions. Approaches like language-conditioned behavioral cloning (LCBC) train policies to directly imitate expert actions conditioned on language, but require humans to annotate all training trajectories and generalize poorly across scenes and behaviors. Meanwhile, recent goal-conditioned approaches perform much better at general manipulation tasks, but do not enable easy task specification for human operators. How can we reconcile the ease of specifying tasks through LCBC-like approaches with the performance improvements of goal-conditioned learning?

TL;DR: In RLHF, there’s tension between the reward learning phase, which uses human preference in the form of comparisons, and the RL fine-tuning phase, which optimizes a single, non-comparative reward. What if we performed RL in a comparative way?

Figure 1:
This diagram illustrates the difference between reinforcement learning from absolute feedback and relative feedback. By incorporating a new component – pairwise policy gradient, we can unify the reward modeling stage and RL stage, enabling direct updates based on pairwise responses.

Large Language Models (LLMs) have powered increasingly capable virtual assistants, such as GPT-4, Claude-2, Bard and Bing Chat. These systems can respond to complex user queries, write code, and even produce poetry. The technique underlying these amazing virtual assistants is Reinforcement Learning with Human Feedback (RLHF). RLHF aims to align the model with human values and eliminate unintended behaviors, which can often arise due to the model being exposed to a large quantity of low-quality data during its pretraining phase.

Proximal Policy Optimization (PPO), the dominant RL optimizer in this process, has been reported to exhibit instability and implementation complications. More importantly, there’s a persistent discrepancy in the RLHF process: despite the reward model being trained using comparisons between various responses, the RL fine-tuning stage works on individual responses without making any comparisons. This inconsistency can exacerbate issues, especially in the challenging language generation domain.

Given this backdrop, an intriguing question arises: Is it possible to design an RL algorithm that learns in a comparative manner? To explore this, we introduce Pairwise Proximal Policy Optimization (P3O), a method that harmonizes the training processes in both the reward learning stage and RL fine-tuning stage of RLHF, providing a satisfactory solution to this issue.

TL;DR: In RLHF, there’s tension between the reward learning phase, which uses human preference in the form of comparisons, and the RL fine-tuning phase, which optimizes a single, non-comparative reward. What if we performed RL in a comparative way?

Figure 1:
This diagram illustrates the difference between reinforcement learning from absolute feedback and relative feedback. By incorporating a new component – pairwise policy gradient, we can unify the reward modeling stage and RL stage, enabling direct updates based on pairwise responses.

Large Language Models (LLMs) have powered increasingly capable virtual assistants, such as GPT-4, Claude-2, Bard and Bing Chat. These systems can respond to complex user queries, write code, and even produce poetry. The technique underlying these amazing virtual assistants is Reinforcement Learning with Human Feedback (RLHF). RLHF aims to align the model with human values and eliminate unintended behaviors, which can often arise due to the model being exposed to a large quantity of low-quality data during its pretraining phase.

Proximal Policy Optimization (PPO), the dominant RL optimizer in this process, has been reported to exhibit instability and implementation complications. More importantly, there’s a persistent discrepancy in the RLHF process: despite the reward model being trained using comparisons between various responses, the RL fine-tuning stage works on individual responses without making any comparisons. This inconsistency can exacerbate issues, especially in the challenging language generation domain.

Given this backdrop, an intriguing question arises: Is it possible to design an RL algorithm that learns in a comparative manner? To explore this, we introduce Pairwise Proximal Policy Optimization (P3O), a method that harmonizes the training processes in both the reward learning stage and RL fine-tuning stage of RLHF, providing a satisfactory solution to this issue.

Figure 1: stepwise behavior in self-supervised learning. When training common SSL algorithms, we find that the loss descends in a stepwise fashion (top left) and the learned embeddings iteratively increase in dimensionality (bottom left). Direct visualization of embeddings (right; top three PCA directions shown) confirms that embeddings are initially collapsed to a point, which then expands to a 1D manifold, a 2D manifold, and beyond concurrently with steps in the loss.

It is widely believed that deep learning’s stunning success is due in part to its ability to discover and extract useful representations of complex data. Self-supervised learning (SSL) has emerged as a leading framework for learning these representations for images directly from unlabeled data, similar to how LLMs learn representations for language directly from web-scraped text. Yet despite SSL’s key role in state-of-the-art models such as CLIP and MidJourney, fundamental questions like “what are self-supervised image systems really learning?” and “how does that learning actually occur?” lack basic answers.

Our recent paper (to appear at ICML 2023) presents what we suggest is the first compelling mathematical picture of the training process of large-scale SSL methods. Our simplified theoretical model, which we solve exactly, learns aspects of the data in a series of discrete, well-separated steps. We then demonstrate that this behavior can be observed in the wild across many current state-of-the-art systems.
This discovery opens new avenues for improving SSL methods, and enables a whole range of new scientific questions that, when answered, will provide a powerful lens for understanding some of today’s most important deep learning systems.

Figure 1: stepwise behavior in self-supervised learning. When training common SSL algorithms, we find that the loss descends in a stepwise fashion (top left) and the learned embeddings iteratively increase in dimensionality (bottom left). Direct visualization of embeddings (right; top three PCA directions shown) confirms that embeddings are initially collapsed to a point, which then expands to a 1D manifold, a 2D manifold, and beyond concurrently with steps in the loss.

It is widely believed that deep learning’s stunning success is due in part to its ability to discover and extract useful representations of complex data. Self-supervised learning (SSL) has emerged as a leading framework for learning these representations for images directly from unlabeled data, similar to how LLMs learn representations for language directly from web-scraped text. Yet despite SSL’s key role in state-of-the-art models such as CLIP and MidJourney, fundamental questions like “what are self-supervised image systems really learning?” and “how does that learning actually occur?” lack basic answers.

Our recent paper (to appear at ICML 2023) presents what we suggest is the first compelling mathematical picture of the training process of large-scale SSL methods. Our simplified theoretical model, which we solve exactly, learns aspects of the data in a series of discrete, well-separated steps. We then demonstrate that this behavior can be observed in the wild across many current state-of-the-art systems.
This discovery opens new avenues for improving SSL methods, and enables a whole range of new scientific questions that, when answered, will provide a powerful lens for understanding some of today’s most important deep learning systems.

<!– –>
Figure 1: CoarsenConf architecture.

<!– (I) The encoder $q_phi(z| X, mathcal{R})$ takes the fine-grained (FG) ground truth conformer $X$, RDKit approximate conformer $mathcal{R}$ , and coarse-grained (CG) conformer $mathcal{C}$ as inputs (derived from $X$ and a predefined CG strategy), and outputs a variable-length equivariant CG representation via equivariant message passing and point convolutions.
(II) Equivariant MLPs are applied to learn the mean and log variance of both the posterior and prior distributions.
(III) The posterior (training) or prior (inference) is sampled and fed into the Channel Selection module, where an attention layer is used to learn the optimal pathway from CG to FG structure.
(IV) Given the FG latent vector and the RDKit approximation, the decoder $p_theta(X |mathcal{R}, z)$ learns to recover the low-energy FG structure through autoregressive equivariant message passing. The entire model can be trained end-to-end by optimizing the KL divergence of latent distributions and reconstruction error of generated conformers. –>

Molecular conformer generation is a fundamental task in computational chemistry. The objective is to predict stable low-energy 3D molecular structures, known as conformers, given the 2D molecule. Accurate molecular conformations are crucial for various applications that depend on precise spatial and geometric qualities, including drug discovery and protein docking.

We introduce CoarsenConf, an SE(3)-equivariant hierarchical variational autoencoder (VAE) that pools information from fine-grain atomic coordinates to a coarse-grain subgraph level representation for efficient autoregressive conformer generation.

<!– –>
Figure 1: CoarsenConf architecture.

<!– (I) The encoder $q_phi(z| X, mathcal{R})$ takes the fine-grained (FG) ground truth conformer $X$, RDKit approximate conformer $mathcal{R}$ , and coarse-grained (CG) conformer $mathcal{C}$ as inputs (derived from $X$ and a predefined CG strategy), and outputs a variable-length equivariant CG representation via equivariant message passing and point convolutions.
(II) Equivariant MLPs are applied to learn the mean and log variance of both the posterior and prior distributions.
(III) The posterior (training) or prior (inference) is sampled and fed into the Channel Selection module, where an attention layer is used to learn the optimal pathway from CG to FG structure.
(IV) Given the FG latent vector and the RDKit approximation, the decoder $p_theta(X |mathcal{R}, z)$ learns to recover the low-energy FG structure through autoregressive equivariant message passing. The entire model can be trained end-to-end by optimizing the KL divergence of latent distributions and reconstruction error of generated conformers. –>

Molecular conformer generation is a fundamental task in computational chemistry. The objective is to predict stable low-energy 3D molecular structures, known as conformers, given the 2D molecule. Accurate molecular conformations are crucial for various applications that depend on precise spatial and geometric qualities, including drug discovery and protein docking.

We introduce CoarsenConf, an SE(3)-equivariant hierarchical variational autoencoder (VAE) that pools information from fine-grain atomic coordinates to a coarse-grain subgraph level representation for efficient autoregressive conformer generation.