ICLR 2023 Papers

Linear regression is a fundamental tool for statistical analysis. This has motivated the development of linear regression methods that also satisfy differential privacy and thus guarantee that the learned model reveals little about any one data point used to construct it. However, existing differentially private solutions assume that the end user can easily specify good data bounds and hyperparameters. Both present significant practical obstacles. In this paper, we study an algorithm which uses the exponential mechanism to select a model with high Tukey depth from a collection of non-private regression models. Given $n$ samples of $d$-dimensional data used to train $m$ models, we construct an efficient analogue using an approximate Tukey depth that runs in time $O(d^2n + dm\log(m))$. We find that this algorithm obtains strong empirical performance in the data-rich setting with no data bounds or hyperparameter selection required.

Low-rank approximation in data streams is a fundamental and significant task in computing science, machine learning and statistics. Multiple streaming algorithms have emerged over years and most of them are inspired by randomized algorithms, more specifically, sketching methods. However, many algorithms are not able to leverage information of data streams and consequently suffer from low accuracy. Existing data-driven methods improve accuracy but the training cost is expensive in practice. In this paper, from a subspace perspective, we propose a tensor-based sketching method for low-rank approximation of data streams. The proposed algorithm fully exploits the structure of data streams and obtains quasi-optimal sketching matrices by performing tensor decomposition on training data. A series of experiments are carried out and show that the proposed tensor-based method can be more accurate and much faster than the previous work.

We consider the estimation of average and counterfactual treatment effects, under two settings: back-door adjustment and front-door adjustment. The goal in both cases is to recover the treatment effect without having an access to a hidden confounder. This objective is attained by first estimating the conditional mean of the desired outcome variable given relevant covariates (the ``first stage" regression), and then taking the (conditional) expectation of this function as a ``second stage" procedure. We propose to compute these conditional expectations directly using a regression function to the learned input features of the first stage, thus avoiding the need for sampling or density estimation. All functions and features (and in particular, the output features in the second stage) are neural networks learned adaptively from data, with the sole requirement that the final layer of the first stage should be linear. The proposed method is shown to converge to the true causal parameter, and outperforms the recent state-of-the-art methods on challenging causal benchmarks, including settings involving high-dimensional image data.

Recent years have witnessed the success of applying seq2seq models to graph parsing tasks, where the outputs are compositionally structured (e.g., a graph or a tree). However, these seq2seq approaches pose a challenge in quantifying the model’s compositional uncertainty on graph structures due to the gap between seq2seq output probability and structural probability on the graph. This work is the first to quantify and evaluate compositional uncertainty for seq2seq graph parsing tasks. First, we proposed a generic, probabilistically interpretable framework that allows correspondences between seq2seq output probability to structural probability on the graph. This framework serves as a powerful medium for quantifying a seq2seq model's compositional uncertainty on graph elements (i.e., nodes or edges). Second, to evaluate uncertainty quality in terms of calibration, we propose a novel metric called Compositional Expected Calibration Error (CECE) which can measure a model’s calibration behavior in predicting graph structures. By a thorough evaluation for compositional uncertainty on three different tasks across ten domains, we demonstrate that CECE is a better reflection for distributional shift compared to vanilla sequence ECE. Finally, we validate the effectiveness of compositional uncertainty considering the task of collaborative semantic parsing, where the model is allowed to send limited subgraphs for human review. The results show that the collaborative performance based on uncertain subgraph selection consistently outperforms random subgraph selection (30% average error reduction rate) and performs comparably to oracle subgraph selection (only 0.33 difference in average prediction error), indicating that compositional uncertainty is an ideal signal for model errors and can benefit various downstream tasks.

Transformer architectures have achieved great success in solving natural language tasks, which learn strong language representations from large-scale unlabeled texts. In this paper, we seek to go further beyond and explore a new logical inductive bias for better language representation learning. Logic reasoning is known as a formal methodology to reach answers from given knowledge and facts. Inspired by such a view, we develop a novel neural architecture named FOLNet (First-Order Logic Network), to encode this new inductive bias. We construct a set of neural logic operators as learnable Horn clauses, which are further forward-chained into a fully differentiable neural architecture (FOLNet). Interestingly, we find that the self-attention module in transformers can be composed by two of our neural logic operators, which probably explains their strong reasoning performance. Our proposed FOLNet has the same input and output interfaces as other pretrained models and thus could be pretrained/finetuned by using similar losses. It also allows FOLNet to be used in a plug-and-play manner when replacing other pretrained models. With our logical inductive bias, the same set of ``logic deduction skills'' learned through pretraining are expected to be equally capable of solving diverse downstream tasks. For this reason, FOLNet learns language representations that have much stronger transfer capabilities. Experimental results on several language understanding tasks show that our pretrained FOLNet model outperforms the existing strong transformer-based approaches.

In this work, we present CogVideo, a 9B-parameter transformer for text-to-video generation. The CogVideo model has been trained by inheriting a pretrained text-to-image model, CogView2, which significantly reduces the training cost and alleviates the problem of scarcity and weak relevance. We also propose a multi-frame-rate training strategy for better aligning text and video clips. CogVideo achieves state-of-the-art performance in machine evaluation and outperforms publicly available models by a large margin in human evaluation. Its codes and model are also publicly available at https://github.com/THUDM/CogVideo.

Many empirical studies have demonstrated the performance benefits of conditional computation in neural networks, including reduced inference time and power consumption. We study the fundamental limits of neural conditional computation from the perspective of memorization capacity. For Rectified Linear Unit (ReLU) networks without conditional computation, it is known that memorizing a collection of $n$ input-output relationships can be accomplished via a neural network with $O(\sqrt{n})$ neurons. Calculating the output of this neural network can be accomplished using $O(\sqrt{n})$ elementary arithmetic operations of additions, multiplications and comparisons for each input. Using a conditional ReLU network, we show that the same task can be accomplished using only $O(\log n)$ operations per input. This represents an almost exponential improvement as compared to networks without conditional computation. We also show that the $\Theta(\log n)$ rate is the best possible. Our achievability result utilizes a general methodology to synthesize a conditional network out of an unconditional network in a computationally-efficient manner, bridging the gap between unconditional and conditional architectures.

Co-design of morphology and control of a Voxel-based Soft Robot (VSR) is challenging due to the notorious bi-level optimization. In this paper, we present a Curriculum-based Co-design (CuCo) method for learning to design and control VSRs through an easy-to-difficult process. Specifically, we expand the design space from a small size to the target size gradually through a predefined curriculum. At each learning stage of the curriculum, we use reinforcement learning to simultaneously train the design policy and the control policy, which is enabled by incorporating the design process into the environment and using differentiable policy representations. The converged morphology and the learned policies from last stage are inherited and then serve as the starting point for the next stage. In empirical studies, we show that CuCo is more efficient in creating larger robots with better performance by reusing the practical design and control patterns learned within each stage, in comparison to prior approaches that learn from scratch in the space of target size.

We develop a new approximation and estimation analysis of deep feed-forward neural networks (FNNs) with the Rectified Linear Unit (ReLU) activation. The functions of interests for the approximation and estimation are assumed to be from Sobolev spaces defined over the $d$-dimensional unit sphere with smoothness index $r>0$. In the regime where $r$ is in the constant order (i.e., $r=\mathcal{O}(1)$), it is shown that at most $d^d$ active parameters are required for getting $d^{-C}$ approximation rate for some constant $C>0$. In contrast, in the regime where the index $r$ grows in the order of $d$ (i.e., $r=\mathcal{O}(d)$) asymptotically, we prove the approximation error decays in the rate $d^{-d^{\beta}}$ with $0<\beta<1$ up to some constant factor independent of $d$. The required number of active parameters in the networks for the approximation increases polynomially in $d$ as $d\rightarrow{\infty}$. In addition to this, it is shown that bound on the excess risk has a $d^d$ factor, when $r=\mathcal{O}(1)$, whereas it has $d^{\mathcal{O}(1)}$ factor, when $r=\mathcal{O}(d)$. We emphasize our findings by making comparisons to the results on approximation and estimation errors of deep ReLU FNN when functions are from Sobolev spaces defined over $d$-dimensional cube. Here, we show that with the current state-of-the-art result, $d^{d}$ factor remain both in the approximation and estimation error, regardless of the order of $r$.

By properly utilizing the learned environment model, model-based reinforcement learning methods can improve the sample efficiency for decision-making problems. Beyond using the learned environment model to train a policy, the success of MCTS-based methods shows that directly incorporating the learned environment model as a planner to make decisions might be more effective. However, when action space is of high dimension and continuous, directly planning according to the learned model is costly and non-trivial. Because of two challenges: (1) the infinite number of candidate actions and (2) the temporal dependency between actions in different timesteps. To address these challenges, inspired by Differential Dynamic Programming (DDP) in optimal control theory, we design a novel Policy Optimization with Model Planning (POMP) algorithm, which incorporates a carefully designed Deep Differential Dynamic Programming (D3P) planner into the model-based RL framework. In D3P planner, (1) to effectively plan in the continuous action space, we construct a locally quadratic programming problem that uses a gradient-based optimization process to replace search. (2) To take the temporal dependency of actions at different timesteps into account, we leverage the updated and latest actions of previous timesteps (i.e., step $1, \cdots, h-1$) to update the action of the current step (i.e., step $h$), instead of updating all actions simultaneously. We theoretically prove the convergence rate for our D3P planner and analyze the effect of the feedback term. In practice, to effectively apply the neural network based D3P planner in reinforcement learning, we leverage the policy network to initialize the action sequence and keep the action update conservative in the planning process. Experiments demonstrate that POMP consistently improves sample efficiency on widely used continuous control tasks. Our code is released at https://github.com/POMP-D3P/POMP-D3P.

Algorithmic case-based decision support provides examples to help human make sense of predicted labels and aid human in decision-making tasks. Despite the promising performance of supervised learning, representations learned by supervised models may not align well with human intuitions: what models consider as similar examples can be perceived as distinct by humans. As a result, they have limited effectiveness in case-based decision support. In this work, we incorporate ideas from metric learning with supervised learning to examine the importance of alignment for effective decision support. In addition to instance-level labels, we use human-provided triplet judgments to learn human-compatible decision-focused representations. Using both synthetic data and human subject experiments in multiple classification tasks, we demonstrate that such representation is better aligned with human perception than representation solely optimized for classification. Human-compatible representations identify nearest neighbors that are perceived as more similar by humans and allow humans to make more accurate predictions, leading to substantial improvements in human decision accuracies (17.8% in butterfly vs. moth classification and 13.2% in pneumonia classification).

Inverse molecular design is critical in material science and drug discovery, where the generated molecules should satisfy certain desirable properties. In this paper, we propose equivariant energy-guided stochastic differential equations (EEGSDE), a flexible framework for controllable 3D molecule generation under the guidance of an energy function in diffusion models. Formally, we show that EEGSDE naturally exploits the geometric symmetry in 3D molecular conformation, as long as the energy function is invariant to orthogonal transformations. Empirically, under the guidance of designed energy functions, EEGSDE significantly improves the baseline on QM9, in inverse molecular design targeted to quantum properties and molecular structures. Furthermore, EEGSDE is able to generate molecules with multiple target properties by combining the corresponding energy functions linearly.

Privacy is a central tenet of Federated learning (FL), in which a central server trains models without centralizing user data. However, gradient updates used in FL can leak user information. While the most industrial uses of FL are for text applications (e.g. keystroke prediction), the majority of attacks on user privacy in FL have focused on simple image classifiers and threat models that assume honest execution of the FL protocol from the server. We propose a novel attack that reveals private user text by deploying malicious parameter vectors, and which succeeds even with mini-batches, multiple users, and long sequences. Unlike previous attacks on FL, the attack exploits characteristics of both the Transformer architecture and the token embedding, separately extracting tokens and positional embeddings to retrieve high-fidelity text. We argue that the threat model of malicious server states is highly relevant from a user-centric perspective, and show that in this scenario, text applications using transformer models are much more vulnerable than previously thought.

Traditional Dictionary Learning (DL) aims to approximate data vectors as sparse linear combinations of basis elements (atoms) and is widely used in machine learning, computer vision, and signal processing. To extend DL to graphs, Vincent-Cuaz et al. 2021 propose a method, called GDL, which describes the topology of each graph with a pairwise relation matrix (PRM) and compares PRMs via the Gromov-Wasserstein Discrepancy (GWD). However, the lack of robustness often excludes GDL from a variety of real-world applications since GWD is sensitive to the structural noise in graphs. This paper proposes an improved graph dictionary learning algorithm based on a robust Gromov-Wasserstein discrepancy (RGWD) which has theoretically sound properties and an efficient numerical scheme. Based on such a discrepancy, our dictionary learning algorithm can learn atoms from noisy graph data. Experimental results demonstrate that our algorithm achieves good performance on both simulated and real-world datasets.

The method of random Fourier features (RFF), proposed in a seminal paper by Rahimi and Recht (NIPS'07), is a powerful technique to find approximate low-dimensional representations of points in (high-dimensional) kernel space, for shift-invariant kernels. While RFF has been analyzed under various notions of error guarantee, the ability to preserve the kernel distance with \emph{relative} error is less understood. We show that for a significant range of kernels, including the well-known Laplacian kernels, RFF cannot approximate the kernel distance with small relative error using low dimensions. We complement this by showing as long as the shift-invariant kernel is analytic, RFF with $\mathrm{poly}(\epsilon^{-1} \log n)$ dimensions achieves $\epsilon$-relative error for pairwise kernel distance of $n$ points, and the dimension bound is improved to $\mathrm{poly}(\epsilon^{-1}\log k)$ for the specific application of kernel $k$-means. Finally, going beyond RFF, we make the first step towards data-oblivious dimension-reduction for general shift-invariant kernels, and we obtain a similar $\mathrm{poly}(\epsilon^{-1} \log n)$ dimension bound for Laplacian kernels. We also validate the dimension-error tradeoff of our methods on simulated datasets, and they demonstrate superior performance compared with other popular methods including random-projection and Nystr\"{o}m methods.

Various saliency map methods have been proposed to interpret and explain predictions of deep learning models. Saliency maps allow us to interpret which parts of the input signals have a strong influence on the prediction results. However, since a saliency map is obtained by complex computations in deep learning models, it is often difficult to know how reliable the saliency map itself is. In this study, we propose a method to quantify the reliability of a saliency region in the form of p-values. Our idea is to consider a saliency map as a selected hypothesis by the trained deep learning model and employ the selective inference framework. The proposed method provably provides a valid p-value for the detected salient region, i.e., we can provably control the false positive rate of the detected salient region. We demonstrate the validity of the proposed method through numerical examples in synthetic and real datasets. Furthermore, we develop a Keras-based framework for conducting the proposed selective inference for a wide class of CNNs without additional implementation cost.

Typical continual learning setup assumes that the dataset is split into multiple discrete tasks. We argue that it is less realistic as the streamed data would have no notion of task boundary in real-world data. Here, we take a step forward to investigate more realistic online continual learning – learning continuously changing data distribution without explicit task boundary, which we call boundary-free setup. As there is no clear boundary of tasks, it is not obvious when and what information in the past to be preserved as a better remedy for the stability-plasticity dilemma. To this end, we propose a scheduled transfer of previously learned knowledge. We further propose a data-driven balancing between the knowledge in the past and the present in learning objective. Moreover, since it is not straight-forward to use the previously proposed forgetting measure without task boundaries, we further propose a novel forgetting measure based on information theory that can capture forgetting. We empirically evaluate our method on a Gaussian data stream, its periodic extension, which assumes periodic data distribution frequently observed in real-life data, as well as the conventional disjoint task-split. Our method outperforms prior arts by large margins in various setups, using four popular benchmark datasets – CIFAR-10, CIFAR-100, TinyImageNet and ImageNet.

We propose a novel \underline{e}dge guided \underline{g}enerative \underline{a}dversarial \underline{n}etwork with \underline{c}ontrastive learning (ECGAN) for the challenging semantic image synthesis task. Although considerable improvement has been achieved, the quality of synthesized images is far from satisfactory due to three largely unresolved challenges. 1) The semantic labels do not provide detailed structural information, making it difficult to synthesize local details and structures. 2) The widely adopted CNN operations such as convolution, down-sampling, and normalization usually cause spatial resolution loss and thus cannot fully preserve the original semantic information, leading to semantically inconsistent results (e.g., missing small objects). 3) Existing semantic image synthesis methods focus on modeling ``local'' semantic information from a single input semantic layout. However, they ignore ``global'' semantic information of multiple input semantic layouts, i.e., semantic cross-relations between pixels across different input layouts. To tackle 1), we propose to use edge as an intermediate representation which is further adopted to guide image generation via a proposed attention guided edge transfer module. Edge information is produced by a convolutional generator and introduces detailed structure information. To tackle 2), we design an effective module to selectively highlight class-dependent feature maps according to the original semantic layout to preserve the semantic information. To tackle 3), inspired by current methods in contrastive learning, we propose a novel contrastive learning method, which aims to enforce pixel embeddings belonging to the same semantic class to generate more similar image content than those from different classes. Doing so can capture more semantic relations by explicitly exploring the structures of labeled pixels from multiple input semantic layouts. Experiments on three challenging datasets show that our ECGAN achieves significantly better results than state-of-the-art methods.

Many recent approaches to natural language tasks are built on the remarkable abilities of large language models. Large language models can perform in-context learning, where they learn a new task from a few task demonstrations, without any parameter updates. This work examines the implications of in-context learning for the creation of datasets for new natural language tasks. Departing from recent in-context learning methods, we formulate an annotation-efficient, two-step framework: selective annotation that chooses a pool of examples to annotate from unlabeled data in advance, followed by prompt retrieval that retrieves task examples from the annotated pool at test time. Based on this framework, we propose an unsupervised, graph-based selective annotation method, voke-k, to select diverse, representative examples to annotate. Extensive experiments on 10 datasets (covering classification, commonsense reasoning, dialogue, and text/code generation) demonstrate that our selective annotation method improves the task performance by a large margin. On average, vote-k achieves a 12.9%/11.4% relative gain under an annotation budget of 18/100, as compared to randomly selecting examples to annotate. Compared to state-of-the-art supervised finetuning approaches, it yields similar performance with 10-100x less annotation cost across 10 tasks. We further analyze the effectiveness of our framework in various scenarios: language models with varying sizes, alternative selective annotation methods, and cases where there is a test data domain shift. We hope that our studies will serve as a basis for data annotations as large language models are increasingly applied to new tasks.