ICML 2024 Papers

The strong performance of neural networks typically hinges on the availability of extensive labeled data, yet acquiring ground-truth labels is often challenging. Instead, noisy supervisions from multiple sources, e.g., by multiple well-designed rules, are more convenient to collect. In this paper, we focus on the realistic problem of learning from multiple noisy label sources, and argue that prior studies have overlooked the crucial *self-cognition* ability of neural networks, i.e., the inherent capability of autonomously distinguishing noise during training. We theoretically analyze this ability of neural networks when meeting multiple noisy label sources, which reveals that neural networks possess the capability to recognize both instance-wise noise within each single noisy label source and annotator-wise quality among multiple noisy label sources. Inspired by the theoretical analyses, we introduce an approach named Self-cognitive Denoising for Multiple noisy label sources (SDM), which exploits the self-cognition ability of neural networks to denoise during training. Furthermore, we build a selective distillation module following the theoretical insights to optimize computational efficiency. The experiments on various datasets demonstrate the superiority of our method.

Model quantization is a compression technique that converts a full-precision model to a more compact low-precision version for better storage. Despite the great success of quantization, recent studies revealed the feasibility of malicious exploiting model quantization via implanting quantization-conditioned backdoors (QCBs). These special backdoors remain dormant in full-precision models but are exposed upon quantization. Unfortunately, existing defenses have limited effects on mitigating QCBs. In this paper, we conduct an in-depth analysis of QCBs. We reveal an intriguing characteristic of QCBs, where activation of backdoor-related neurons on even benign samples enjoy a distribution drift after quantization, although this drift is more significant on poisoned samples. Motivated by this finding, we propose to purify the backdoor-exposed quantized model by aligning its layer-wise activation with its full-precision version. To further exploit the more pronounced activation drifts on poisoned samples, we design an additional module to layer-wisely approximate poisoned activation distribution based on batch normalization statistics of the full-precision model. Extensive experiments are conducted, verifying the effectiveness of our defense. Our code is publicly available.

This paper introduces RaLMSpec, a framework that accelerates iterative retrieval-augmented language model (RaLM) with *speculative retrieval* and *batched verification*. RaLMSpec further introduces several important systems optimizations, including prefetching, optimal speculation stride scheduler, and asynchronous verification. The combination of these techniques allows RaLMSPec to significantly outperform existing systems. For document-level iterative RaLM serving, evaluation over three LLMs on four QA datasets shows that RaLMSpec improves over existing approaches by $1.75$-$2.39\times$, $1.04$-$1.39\times$, and $1.31$-$1.77\times$ when the retriever is an exact dense retriever, approximate dense retriever, and sparse retriever respectively. For token-level iterative RaLM (KNN-LM) serving, RaLMSpec is up to $7.59\times$ and $2.45\times$ faster than existing methods for exact dense and approximate dense retrievers, respectively.

Rooting in the scarcity of most attributes, realistic pedestrian attribute datasets exhibit unduly skewed data distribution, from which two types of model failures are delivered: (1) label imbalance: model predictions lean greatly towards the side of majority labels; (2) semantics imbalance: model is easily overfitted on the under-represented attributes due to their insufficient semantic diversity. To render perfect label balancing, we propose a novel framework that successfully decouples label-balanced data re-sampling from the curse of attributes co-occurrence, i.e., we equalize the sampling prior of an attribute while not biasing that of the co-occurred others. To diversify the attributes semantics and mitigate the feature noise, we propose a Bayesian feature augmentation method to introduce true in-distribution novelty. Handling both imbalances jointly, our work achieves best accuracy on various popular benchmarks, and importantly, with minimal computational budget.

Fast adapting to unknown peers (partners or opponents) with different strategies is a key challenge in multi-agent games. To do so, it is crucial for the agent to probe and identify the peer’s strategy efficiently, as this is the prerequisite for carrying out the best response in adaptation. However, exploring the strategies of unknown peers is difficult, especially when the games are partially observable and have a long horizon. In this paper, we propose a peer identification reward, which rewards the learning agent based on how well it can identify the behavior pattern of the peer over the historical context, such as the observation over multiple episodes. This reward motivates the agent to learn a context-aware policy for effective exploration and fast adaptation, i.e., to actively seek and collect informative feedback from peers when uncertain about their policies and to exploit the context to perform the best response when confident. We evaluate our method on diverse testbeds that involve competitive (Kuhn Poker), cooperative (PO-Overcooked), or mixed (Predator-Prey-W) games with peer agents. We demonstrate that our method induces more active exploration behavior, achieving faster adaptation and better outcomes than existing methods.

In several real-world sequential decision problems, at every step, the learner is required to select different actions. Every action affects a specific part of the system and generates an observable intermediate effect. In this paper, we introduce the Factored-Reward Bandits (FRBs), a novel setting able to effectively capture and exploit the structure of this class of scenarios, where the reward is computed as the product of the action intermediate observations. We characterize the statistical complexity of the learning problem in the FRBs, by deriving worst-case and asymptotic instance-dependent regret lower bounds. Then, we devise and analyze two regret minimization algorithms. The former, F-UCB, is an anytime optimistic approach matching the worst-case lower bound (up to logarithmic factors) but fails to perform optimally from the instance-dependent perspective. The latter, F-Track, is a bound-tracking approach, that enjoys optimal asymptotic instance-dependent regret guarantees.

Exploration is a major challenge in reinforcement learning, especially for high-dimensional domains that require function approximation. We propose exploration objectives---policy optimization objectives that enable downstream maximization of any reward function---as a conceptual framework to systematize the study of exploration. We introduce a new objective, L1-Coverage, which generalizes previous exploration schemes and supports three fundamental desiderata: 1. *Intrinsic complexity control.* L1-Coverage is associated with a structural parameter, L1-Coverability, which reflects the intrinsic statistical difficulty of the underlying MDP, subsuming Block and Low-Rank MDPs. 2. *Efficient planning.* For a known MDP, L1-Coverage efficiently reduces to standard policy optimization, allowing flexible integration with off-the-shelf methods such as policy gradient and Q-learning approaches. 3. *Efficient exploration.* L1-Coverage enables the first computationally efficient model-based and model-free algorithms for online (reward-free or reward-driven) reinforcement learning in MDPs with low coverability. Empirically, we find that L1-Coverage effectively drives off-the-shelf policy optimization algorithms to explore the state space.

In contrast to the prevailing interpretation of Low-Rank Adaptation (LoRA) as a means of simulating weight changes in model adaptation, this paper introduces an alternative perspective by framing it as a control process. Specifically, we conceptualize lightweight matrices in LoRA as control modules tasked with perturbing the original, complex, yet frozen blocks on downstream tasks. Building upon this new understanding, we conduct a thorough analysis on the controllability of these modules, where we identify and establish sufficient conditions that facilitate their effective integration into downstream controls. Moreover, the control modules are redesigned by incorporating nonlinearities through a parameter-free attention mechanism. This modification allows for the intermingling of tokens within the controllers, enhancing the adaptability and performance of the system. Empirical findings substantiate that, without introducing any additional parameters, this approach surpasses the existing LoRA algorithms across all assessed datasets and rank configurations.

Visual Model-Based Reinforcement Learning (MBRL) promises to encapsulate agent's knowledge about the underlying dynamics of the environment, enabling learning a world model as a useful planner. However, top MBRL agents such as Dreamer often struggle with visual pixel-based inputs in the presence of exogenous or irrelevant noise in the observation space, due to failure to capture task-specific features while filtering out irrelevant spatio-temporal details. To tackle this problem, we apply a spatio-temporal masking strategy, a bisimulation principle, combined with latent reconstruction, to capture endogenous task-specific aspects of the environment for world models, effectively eliminating non-essential information. Joint training of representations, dynamics, and policy often leads to instabilities. To further address this issue, we develop a Hybrid Recurrent State-Space Model (HRSSM) structure, enhancing state representation robustness for effective policy learning. Our empirical evaluation demonstrates significant performance improvements over existing methods in a range of visually complex control tasks such as Maniskill with exogenous distractors from the Matterport environment. Our code is avaliable at https://github.com/bit1029public/HRSSM.

Recent works like Tree-of-Thought (ToT) and Reasoning via Planning (RAP) aim to augment the multi-step reasoning capabilities of LLMs by using tree-search algorithms. These methods rely on prompting a pre-trained model to serve as a value function and focus on problems with low search depth. As a result, these methods cannot benefit from in-domain training and only rely on pretraining process — they will not work in domains where the pre-trained LLM does not have enough knowledge to serve as an effective value function or in domains that require long-horizon planning. To address these limitations, we present an AlphaZero-like tree-search learning framework for LLMs (termed TS-LLM), systematically illustrating how tree-search with a learned value function can guide LLM decoding. TS-LLM distinguishes itself in two key ways. (1) Leveraging a learned value function and AlphaZero-like algorithms, our approach can be generally adaptable to a wide range of tasks, language models of any size, and tasks of varying search depths. (2) Our approach can guide LLMs during both inference and training, iteratively improving the LLMs. Empirical results across reasoning, planning, alignment, and decision-making tasks show that TS-LLM outperforms existing approaches and can handle trees with a depth of 64.

We study the generic identifiability of causal effects in linear non-Gaussian acyclic models (LiNGAM) with latent variables. We consider the problem in two main settings: When the causal graph is known a priori, and when it is unknown. In both settings, we provide a complete graphical characterization of the identifiable direct or total causal effects among observed variables. Moreover, we propose efficient algorithms to certify the graphical conditions. Finally, we propose an adaptation of the reconstruction independent component analysis (RICA) algorithm that estimates the causal effects from the observational data given the causal graph. Experimental results show the effectiveness of the proposed method in estimating the causal effects.

Despite the great progress of 3D vision, data privacy and security issues in 3D deep learning are not explored systematically. In the domain of 2D images, many availability attacks have been proposed to prevent data from being illicitly learned by unauthorized deep models. However, unlike images represented on a fixed dimensional grid, point clouds are characterized as unordered and unstructured sets, posing a significant challenge in designing an effective availability attack for 3D deep learning. In this paper, we theoretically show that extending 2D availability attacks directly to 3D point clouds under distance regularization is susceptible to the degeneracy, rendering the generated poisons weaker or even ineffective. This is because in bi-level optimization, introducing regularization term can result in update directions out of control. To address this issue, we propose a novel Feature Collision Error-Minimization (FC-EM) method, which creates additional shortcuts in the feature space, inducing different update directions to prevent the degeneracy of bi-level optimization. Moreover, we provide a theoretical analysis that demonstrates the effectiveness of the FC-EM attack. Extensive experiments on typical point cloud datasets, 3D intracranial aneurysm medical dataset, and 3D face dataset verify the superiority and practicality of our approach.

Preference-based Reinforcement Learning (PbRL) circumvents the need for reward engineering by harnessing human preferences as the reward signal. However, current PbRL methods excessively depend on high-quality feedback from domain experts, which results in a lack of robustness. In this paper, we present RIME, a robust PbRL algorithm for effective reward learning from noisy preferences. Our method utilizes a sample selection-based discriminator to dynamically filter out noise and ensure robust training. To counteract the cumulative error stemming from incorrect selection, we suggest a warm start for the reward model, which additionally bridges the performance gap during the transition from pre-training to online training in PbRL. Our experiments on robotic manipulation and locomotion tasks demonstrate that RIME significantly enhances the robustness of the state-of-the-art PbRL method. Code is available at https://github.com/CJReinforce/RIME_ICML2024.

Heuristics are widely used for dealing with complex search and optimization problems. However, manual design of heuristics can be often very labour extensive and requires rich working experience and knowledge. This paper proposes Evolution of Heuristic (EoH), a novel evolutionary paradigm that leverages both Large Language Models (LLMs) and Evolutionary Computation (EC) methods for Automatic Heuristic Design (AHD). EoH represents the ideas of heuristics in natural language, termed thoughts. They are then translated into executable codes by LLMs. The evolution of both thoughts and codes in an evolutionary search framework makes it very effective and efficient for generating high-performance heuristics. Experiments on three widely studied combinatorial optimization benchmark problems demonstrate that EoH outperforms commonly used handcrafted heuristics and other recent AHD methods including FunSearch. Particularly, the heuristic produced by EoH with a low computational budget (in terms of the number of queries to LLMs) significantly outperforms widely-used human hand-crafted baseline algorithms for the online bin packing problem.

Mastering multiple tasks through exploration and learning in an environment poses a significant challenge in reinforcement learning (RL). Unsupervised RL has been introduced to address this challenge by training policies with intrinsic rewards rather than extrinsic rewards. However, current intrinsic reward designs and unsupervised RL algorithms often overlook the heterogeneous nature of collected samples, thereby diminishing their sample efficiency. To overcome this limitation, in this paper, we proposed a reward-free RL algorithm called GFA-RFE. The key idea behind our algorithm is an uncertainty-aware intrinsic reward for exploring the environment and an uncertainty-weighted learning process to handle heterogeneous uncertainty in different samples. Theoretically, we show that in order to find an $\epsilon$-optimal policy, GFA-RFE needs to collect $\tilde{O} (H^2 \log N_{\mathcal{F}} (\epsilon) \text{dim} (\mathcal{F}) / \epsilon^2 )$ number of episodes, where $\mathcal{F}$ is the value function class with covering number $N_{\mathcal{F}} (\epsilon)$ and generalized eluder dimension $\text{dim} (\mathcal{F})$. Such a result outperforms all existing reward-free RL algorithms. We further implement and evaluate GFA-RFE across various domains and tasks in the DeepMind Control Suite. Experiment results show that GFA-RFE outperforms or is comparable to the performance of state-of-the-art unsupervised RL algorithms.

This paper explores the connections between tempering (for Sequential Monte Carlo; SMC) and entropic mirror descent to sample from a target probability distribution whose unnormalized density is known. We establish that tempering SMC corresponds to entropic mirror descent applied to the reverse Kullback-Leibler (KL) divergence and obtain convergence rates for the tempering iterates. Our result motivates the tempering iterates from an optimization point of view, showing that tempering can be seen as a descent scheme of the KL divergence with respect to the Fisher-Rao geometry, in contrast to Langevin dynamics that perform descent of the KL with respect to the Wasserstein-2 geometry. We exploit the connection between tempering and mirror descent iterates to justify common practices in SMC and derive adaptive tempering rules that improve over other alternative benchmarks in the literature.

Developing policies that can adapt to non-stationary environments is essential for real-world reinforcement learning applications. Nevertheless, learning such adaptable policies in offline settings, with only a limited set of pre-collected trajectories, presents significant challenges. A key difficulty arises because the limited offline data makes it hard for the context encoder to differentiate between changes in the environment dynamics and shifts in the behavior policy, often leading to context misassociations. To address this issue, we introduce a novel approach called debiased offline representation learning for fast online adaptation (DORA). DORA incorporates an information bottleneck principle that maximizes mutual information between the dynamics encoding and the environmental data, while minimizing mutual information between the dynamics encoding and the actions of the behavior policy. We present a practical implementation of DORA, leveraging tractable bounds of the information bottleneck principle. Our experimental evaluation across six benchmark MuJoCo tasks with variable parameters demonstrates that DORA not only achieves a more precise dynamics encoding but also significantly outperforms existing baselines in terms of performance.

We propose a novel algorithm that extends the methods of ball smoothing and Gaussian smoothing for noisy derivative-free optimization by accounting for the heterogeneous curvature of the objective function. The algorithm dynamically adapts the shape of the smoothing kernel to approximate the Hessian of the objective function around a local optimum. This approach significantly reduces the error in estimating the gradient from noisy evaluations through sampling. We demonstrate the efficacy of our method through numerical experiments on artificial problems. Additionally, we show improved performance when tuning NP-hard combinatorial optimization solvers compared to existing state-ofthe-art heuristic derivative-free and Bayesian optimization methods.

The ever-growing ecosystem of LLMs has posed a challenge in selecting the most appropriate pre-trained model to fine-tune amidst a sea of options. Given constrained resources, fine-tuning all models and making selections afterward is unrealistic. In this work, we formulate this resource-constrained selection task into predicting fine-tuning performance and illustrate its natural connection with Scaling Law. Unlike pre-training, we find that the fine-tuning scaling curve includes not just the well-known "power phase" but also the previously unobserved "pre-power phase". We also explain why existing Scaling Law fails to capture this phase transition phenomenon both theoretically and empirically. To address this, we introduce the concept of "pre-learned data size" into our Rectified Scaling Law, which overcomes theoretical limitations and fits experimental results much better. By leveraging our law, we propose a novel LLM selection algorithm that selects the near-optimal model with hundreds of times less resource consumption, while other methods may provide negatively correlated selection. The project page is available at rectified-scaling-law.github.io.