We study generalizable policy learning from demonstrations for complex low-level control (e.g., contact-rich object manipulations). We propose a novel hierarchical imitation learning method that utilizes sub-optimal demos. Firstly, we propose an observation space-agnostic approach that efficiently discovers the multi-step subskill decomposition of the demos in an unsupervised manner. By grouping temporarily close and functionally similar actions into subskill-level demo segments, the observations at the segment boundaries constitute a chain of planning steps for the task, which we refer to as the chain-of-thought (CoT). Next, we propose a Transformer-based design that effectively learns to predict the CoT as the subskill-level guidance. We couple action and subskill predictions via learnable prompt tokens and a hybrid masking strategy, which enable dynamically updated guidance at test time and improve feature representation of the trajectory for generalizable policy learning. Our method, Chain-of-Thought Predictive Control (CoTPC), consistently surpasses existing strong baselines on various challenging low-level manipulation tasks with sub-optimal demos. See project page at https://sites.google.com/view/cotpc.
论文检索
输入标题、作者或关键词，从 2,778 篇学术成果中精准定位
Applications · Computer Vision
Cross-modal transformers have demonstrated superiority in various vision tasks by effectively integrating different modalities. This paper first critiques prior token exchange methods which replace less informative tokens with inter-modal features, and demonstrate exchange based methods underperform cross-attention mechanisms, while the computational demand of the latter inevitably restricts its use with longer sequences. To surmount the computational challenges, we propose *GeminiFusion*, a pixel-wise fusion approach that capitalizes on aligned cross-modal representations. *GeminiFusion* elegantly combines intra-modal and inter-modal attentions, dynamically integrating complementary information across modalities. We employ a layer-adaptive noise to adaptively control their interplay on a per-layer basis, thereby achieving a harmonized fusion process. Notably, *GeminiFusion* maintains linear complexity with respect to the number of input tokens, ensuring this multimodal framework operates with efficiency comparable to unimodal networks. Comprehensive evaluations across multimodal image-to-image translation, $3$D object detection and arbitrary-modal semantic segmentation tasks, including RGB, depth, LiDAR, event data, etc. demonstrate the superior performance of our *GeminiFusion* against leading-edge techniques. The PyTorch code is available [here](https://github.com/JiaDingCN/GeminiFusion).
Probabilistic Methods · Bayesian Models and Methods
We present Neural Quantile Estimation (NQE), a novel Simulation-Based Inference (SBI) method based on conditional quantile regression. NQE autoregressively learns individual one dimensional quantiles for each posterior dimension, conditioned on the data and previous posterior dimensions. Posterior samples are obtained by interpolating the predicted quantiles using monotonic cubic Hermite spline, with specific treatment for the tail behavior and multi-modal distributions. We introduce an alternative definition for the Bayesian credible region using the local Cumulative Density Function (CDF), offering substantially faster evaluation than the traditional Highest Posterior Density Region (HPDR). In case of limited simulation budget and/or known model misspecification, a post-processing calibration step can be integrated into NQE to ensure the unbiasedness of the posterior estimation with negligible additional computational cost. We demonstrate that NQE achieves state-of-the-art performance on a variety of benchmark problems.
Applications · Computer Vision
In this paper, we address the challenge of Perspective-Invariant Learning in machine learning and computer vision, which involves enabling a network to understand images from varying perspectives to achieve consistent semantic interpretation. While standard approaches rely on the labor-intensive collection of multi-view images or limited data augmentation techniques, we propose a novel framework, Discrete Latent Perspective Learning (DLPL), for latent multi-perspective fusion learning using conventional single-view images. DLPL comprises three main modules: Perspective Discrete Decomposition (PDD), Perspective Homography Transformation (PHT), and Perspective Invariant Attention (PIA), which work together to discretize visual features, transform perspectives, and fuse multi-perspective semantic information, respectively. DLPL is a universal perspective learning framework applicable to a variety of scenarios and vision tasks. Extensive experiments demonstrate that DLPL significantly enhances the network's capacity to depict images across diverse scenarios (daily photos, UAV, auto-driving) and tasks (detection, segmentation).
Reinforcement Learning · Online
Learning high-quality $Q$-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works primarily focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that $Q$-values are often underestimated in the latter stage of the RL training process, potentially hindering policy learning and reducing sample efficiency. We find that such a long-neglected phenomenon is often related to the use of inferior actions from the current policy in Bellman updates as compared to the more optimal action samples in the replay buffer. We propose the Blended Exploitation and Exploration (BEE) operator, a simple yet effective approach that updates $Q$-value using both historical best-performing actions and the current policy. Based on BEE, the resulting practical algorithm BAC outperforms state-of-the-art methods in **over 50** continuous control tasks and achieves strong performance in failure-prone scenarios and **real-world robot** tasks. Benchmark results and videos are available at https://jity16.github.io/BEE/.
Deep Learning · Large Language Models
The alignment of language models with human preferences is vital for their application in real-world tasks. The problem is formulated as optimizing the model's policy to maximize the expected reward that reflects human preferences with minimal deviation from the initial policy. While considered as a straightforward solution, reinforcement learning (RL) suffers from high variance in policy updates, which impedes efficient policy improvement. Recently, direct preference optimization (DPO) was proposed to directly optimize the policy from preference data. However, we show that DPO derived based on the optimal solution of the problem leads to a compromised mean-seeking approximation of the optimal solution in practice. In this paper, we propose efficient exact optimization (EXO) of the alignment objective. EXO is guaranteed to optimize in the same direction as RL algorithms asymptotically for arbitrary policy parametrization. This leads to the same mode-seeking solution, while enables efficient optimization by circumventing the complexities of RL. We also compare our method to DPO with both theoretical and empirical analyses, and further demonstrate the advantages of our method over existing approaches on realistic human preference data. Code is available at https://github.com/haozheji/exact-optimization.
Reinforcement Learning
The varying significance of distinct primitive behaviors during the policy learning process has been overlooked by prior model-free RL algorithms. Leveraging this insight, we explore the causal relationship between different action dimensions and rewards to evaluate the significance of various primitive behaviors during training. We introduce a causality-aware entropy term that effectively identifies and prioritizes actions with high potential impacts for efficient exploration. Furthermore, to prevent excessive focus on specific primitive behaviors, we analyze the gradient dormancy phenomenon and introduce a dormancy-guided reset mechanism to further enhance the efficacy of our method. Our proposed algorithm, **ACE**: Off-policy **A**ctor-critic with **C**ausality-aware **E**ntropy regularization, demonstrates a substantial performance advantage across 29 diverse continuous control tasks spanning 7 domains compared to model-free RL baselines, which underscores the effectiveness, versatility, and efficient sample efficiency of our approach. Benchmark results and videos are available at https://ace-rl.github.io/.
Dynamic Sparse Training (DST) is an effective approach for addressing the substantial training resource requirements posed by the ever-increasing size of the Deep Neural Networks (DNNs). Characterized by its dynamic "train-prune-grow'' schedule during training, DST implicitly develops a bi-level structure for training the weights while discovering a subnetwork topology. However, such a structure is consistently overlooked by the current DST algorithms for further optimization opportunities, and these algorithms, on the other hand, solely optimize the weights while determining masks heuristically. In this paper, we extensively study DST algorithms and argue that the training scheme of DST naturally forms a bi-level problem in which the updating of weight and mask is interdependent. Based on this observation, we introduce a novel efficient training framework called BiDST, which for the first time, introduces bi-level optimization methodology into dynamic sparse training domain. Unlike traditional partial-heuristic DST schemes, which suffer from sub-optimal search efficiency for masks and miss the opportunity to fully explore the topological space of neural networks, BiDST excels at discovering excellent sparse patterns by optimizing mask and weight simultaneously, resulting in maximum 2.62% higher accuracy, 2.1$\times$ faster execution speed, and 25$\times$ reduced overhead. Code available at https://github.com/jjsrf/BiDST-ICML2024.
Reinforcement Learning
This paper proposes a step toward approximate Bayesian inference in on-policy actor-critic deep reinforcement learning. It is implemented through three changes to the Asynchronous Advantage Actor-Critic (A3C) algorithm: (1) applying a ReLU function to advantage estimates, (2) spectral normalization of actor-critic weights, and (3) incorporating *dropout as a Bayesian approximation*. We prove under standard assumptions that restricting policy updates to positive advantages optimizes for value by maximizing a lower bound on the value function plus an additive term. We show that the additive term is bounded proportional to the Lipschitz constant of the value function, which offers theoretical grounding for spectral normalization of critic weights. Finally, our application of dropout corresponds to approximate Bayesian inference over both the actor and critic parameters, which enables *adaptive state-aware* exploration around the modes of the actor via Thompson sampling. We demonstrate significant improvements for median and interquartile mean metrics over A3C, PPO, SAC, and TD3 on the MuJoCo continuous control benchmark and improvement over PPO in the challenging ProcGen generalization benchmark.
Finite Volume Features, Global Geometry Representations, and Residual Training for Deep Learning-based CFD Simulation
PDF ↗Applications · Physics
Computational fluid dynamics (CFD) simulation is an irreplaceable modelling step in many engineering designs, but it is often computationally expensive. Some graph neural network (GNN)-based CFD methods have been proposed. However, the current methods inherit the weakness of traditional numerical simulators, as well as ignore the cell characteristics in the mesh used in the finite volume method, a common method in practical CFD applications. Specifically, the input nodes in these GNN methods have very limited information about any object immersed in the simulation domain and its surrounding environment. Also, the cell characteristics of the mesh such as cell volume, face surface area, and face centroid are not included in the message-passing operations in the GNN methods. To address these weaknesses, this work proposes two novel geometric representations: Shortest Vector (SV) and Directional Integrated Distance (DID). Extracted from the mesh, the SV and DID provide global geometry perspective to each input node, thus removing the need to collect this information through message-passing. This work also introduces the use of Finite Volume Features (FVF) in the graph convolutions as node and edge attributes, enabling its message-passing operations to adjust to different nodes. Finally, this work is the first to demonstrate how residual training, with the availability of low-resolution data, can be adopted to improve the flow field prediction accuracy. Experimental results on two datasets with five different state-of-the-art GNN methods for CFD indicate that SV, DID, FVF and residual training can effectively reduce the predictive error of current GNN-based methods by as much as 41%. Our codes and datasets are available at https://github.com/toggled/FvFGeo.
Deep Learning · Attention Mechanisms
Transformers are the dominant architecture for sequence modeling, but there is growing interest in models that use a fixed-size latent state that does not depend on the sequence length, which we refer to as ''generalized state space models'' (GSSMs). In this paper we show that while GSSMs are promising in terms of inference-time efficiency, they are limited compared to transformer models on tasks that require copying from the input context. We start with a theoretical analysis of the simple task of string copying and prove that a two layer transformer can copy strings of exponential length while GSSMs are fundamentally limited by their fixed-size latent state. Empirically, we find that transformers outperform GSSMs in terms of efficiency and generalization on synthetic tasks that require copying the context. Finally, we evaluate pretrained large language models and find that transformer models dramatically outperform state space models at copying and retrieving information from context. Taken together, these results suggest a fundamental gap between transformers and GSSMs on tasks of practical interest.
Miscellaneous Aspects of Machine Learning · Representation Learning
Domain incremental learning (DIL) poses a significant challenge in real-world scenarios, as models need to be sequentially trained on diverse domains over time, all the while avoiding catastrophic forgetting. Mitigating representation drift, which refers to the phenomenon of learned representations undergoing changes as the model adapts to new tasks, can help alleviate catastrophic forgetting. In this study, we propose a novel DIL method named *DARE*, featuring a three-stage training process: Divergence, Adaptation, and REfinement. This process gradually adapts the representations associated with new tasks into the feature space spanned by samples from previous tasks, simultaneously integrating task-specific decision boundaries. Additionally, we introduce a novel strategy for buffer sampling and demonstrate the effectiveness of our proposed method, combined with this sampling strategy, in reducing representation drift within the feature encoder. This contribution effectively alleviates catastrophic forgetting across multiple DIL benchmarks. Furthermore, our approach prevents sudden representation drift at task boundaries, resulting in a well-calibrated DIL model that maintains the performance on previous tasks.
Deep Learning · Generative Models and Autoencoders
Music generation schemes using language modeling rely on a vocabulary of audio tokens, generally provided as codes in a discrete latent space learnt by an auto-encoder. Multi-stage quantizers are often employed to produce these tokens, therefore the decoding strategy used for token prediction must be adapted to account for multiple codebooks: either it should model the joint distribution over all codebooks, or fit the product of the codebook marginal distributions. Modelling the joint distribution requires a costly increase in the number of auto-regressive steps, while fitting the product of the marginals yields an inexact model unless the codebooks are mutually independent. In this work, we introduce an independence-promoting loss to regularize the auto-encoder used as the tokenizer in language models for music generation. The proposed loss is a proxy for mutual information based on the maximum mean discrepancy principle, applied in reproducible kernel Hilbert spaces. Our criterion is simple to implement and train, and it is generalizable to other multi-stream codecs. We show that it reduces the statistical dependence between codebooks during auto-encoding. This leads to an increase in the generated music quality when modelling the product of the marginal distributions, while generating audio much faster than the joint distribution model.
Social Aspects · Privacy-preserving Statistics and Machine Learning
Deep neural networks (DNNs) have a risk of remembering sensitive data from their training datasets, inadvertently leading to substantial information leakage through privacy attacks like membership inference attacks. DP-SGD is a simple but effective defense method, incorporating Gaussian noise into gradient updates to safeguard sensitive information. With the prevalence of large neural networks, DP-signSGD, a variant of DP-SGD, has emerged, aiming to curtail memory usage while maintaining security. However, it is noteworthy that most DP-signSGD algorithms default to Gaussian noise, suitable only for DP-SGD, without scant discussion of its appropriateness for signSGD. Our study delves into an intriguing question: **"Can we find a more efficient substitute for Gaussian noise to secure privacy in DP-signSGD?"** We propose an answer with a Logistic mechanism, which conforms to signSGD principles and is interestingly evolved from an exponential mechanism. In this paper, we provide both theoretical and experimental evidence showing that our method surpasses DP-signSGD.
Theory · Online Learning and Bandits
Online control with non-stochastic disturbances and adversarially chosen convex cost functions, referred to as online non-stochastic control, has recently attracted increasing attention. We study online non-stochastic control with partial feedback, where learners can only access partially observed states and partially informed (bandit) costs. The problem setting arises naturally in real-world decision-making applications and strictly generalizes exceptional cases studied disparately by previous works. We propose the first online algorithm for this problem, with an $\tilde{O}(T^{3/4})$ regret competing with the best policy in hindsight, where $T$ denotes the time horizon and the $\tilde{O}(\cdot)$-notation omits the poly-logarithmic factors in $T$. To further enhance the algorithms' robustness to changing environments, we then design a novel method with a two-layer structure to optimize the dynamic regret, a more challenging measure that competes with time-varying policies. Our method is based on the online ensemble framework by treating the controller above as the base learner. On top of that, we design two different meta-combiners to simultaneously handle the unknown variation of environments and the memory issue arising from the online control. We prove that the two resulting algorithms enjoy $\tilde{O}(T^{3/4}(1+P_T)^{1/2})$ and $\tilde{O}(T^{3/4}(1+P_T)^{1/4}+T^{5/6})$ dynamic regret respectively, where $P_T$ measures the environmental non-stationarity. Our results are further extended to unknown transition matrices. Finally, empirical studies in both synthetic linear and simulated nonlinear tasks validate our method's effectiveness, thus supporting the theoretical findings.
Reinforcement Learning · Deep RL
We present a novel perspective on goal-conditioned reinforcement learning by framing it within the context of denoising diffusion models. Analogous to the diffusion process, where Gaussian noise is used to create random trajectories that walk away from the data manifold, we construct trajectories that move away from potential goal states. We then learn a goal-conditioned policy to reverse these deviations, analogous to the score function. This approach, which we call Merlin, can reach specified goals from arbitrary initial states without learning a separate value function. In contrast to recent works utilizing diffusion models in offline RL, Merlin stands out as the first method to perform diffusion in the state space, requiring only one "denoising" iteration per environment step. We experimentally validate our approach in various offline goal-reaching tasks, demonstrating substantial performance enhancements compared to state-of-the-art methods while improving computational efficiency over other diffusion-based RL methods by an order of magnitude. Our results suggest that this perspective on diffusion for RL is a simple and scalable approach for sequential decision making.
Social Aspects · Accountability, Transparency and Interpretability
Different users of machine learning methods require different explanations, depending on their goals. To make machine learning accountable to society, one important goal is to get actionable options for recourse, which allow an affected user to change the decision f(x) of a machine learning system by making limited changes to its input x. We formalize this by providing a general definition of recourse sensitivity, which needs to be instantiated with a utility function that describes which changes to the decisions are relevant to the user. This definition applies to local attribution methods, which attribute an importance weight to each input feature. It is often argued that such local attributions should be robust, in the sense that a small change in the input x that is being explained, should not cause a large change in the feature weights. However, we prove formally that it is in general impossible for any single attribution method to be both recourse sensitive and robust at the same time. It follows that there must always exist counterexamples to at least one of these properties. We provide such counterexamples for several popular attribution methods, including LIME, SHAP, Integrated Gradients and SmoothGrad. Our results also cover counterfactual explanations, which may be viewed as attributions that describe a perturbation of x. We further discuss possible ways to work around our impossibility result, for instance by allowing the output to consist of sets with multiple attributions, and we provide sufficient conditions for specific classes of continuous functions to be recourse sensitive. Finally, we strengthen our impossibility result for the restricted case where users are only able to change a single attribute of x, by providing an exact characterization of the functions f to which impossibility applies.
Probabilistic Methods · Bayesian Models and Methods
We propose a new procedure named PASOA, for Bayesian experimental design, that performs sequential design optimization by simultaneously providing accurate estimates of successive posterior distributions for parameter inference. The sequential design process is carried out via a contrastive estimation principle, using stochastic optimization and Sequential Monte Carlo (SMC) samplers to maximise the Expected Information Gain (EIG). As larger information gains are obtained for larger distances between successive posterior distributions, this EIG objective may worsen classical SMC performance. To handle this issue, tempering is proposed to have both a large information gain and an accurate SMC sampling, that we show is crucial for performance. This novel combination of stochastic optimization and tempered SMC allows to jointly handle design optimization and parameter inference. We provide a proof that the obtained optimal design estimators benefit from some consistency property. Numerical experiments confirm the potential of the approach, which outperforms other recent existing procedures.
Reinforcement Learning
Although reinforcement learning (RL) can solve many challenging sequential decision making problems, achieving *zero-shot* transfer across related tasks remains a challenge. The difficulty lies in finding a good representation for the current task so that the agent understands how it relates to previously seen tasks. To achieve zero-shot transfer, we introduce the *function encoder*, a representation learning algorithm which represents a function as a weighted combination of learned, non-linear basis functions. By using a function encoder to represent the reward function or the transition function, the agent has information on how the current task relates to previously seen tasks via a coherent vector representation. Thus, the agent is able to achieve transfer between related tasks at run time with no additional training. We demonstrate state-of-the-art data efficiency, asymptotic performance, and training stability in three RL fields by augmenting basic RL algorithms with a function encoder task representation.
Social Aspects · Trustworthy Machine Learning
Aligning large language models (LLMs) with human intentions has become a critical task for safely deploying models in real-world systems. While existing alignment approaches have seen empirical success, theoretically understanding how these methods affect model behavior remains an open question. Our work provides an initial attempt to theoretically analyze the learning dynamics of human preference alignment. We formally show how the distribution of preference datasets influences the rate of model updates and provide rigorous guarantees on the training accuracy. Our theory also reveals an intricate phenomenon where the optimization is prone to prioritizing certain behaviors with higher preference distinguishability. We empirically validate our findings on contemporary LLMs and alignment tasks, reinforcing our theoretical insights and shedding light on considerations for future alignment approaches. Disclaimer: This paper contains potentially offensive text; reader discretion is advised.