United States

Neural networks are often over-parameterized and hence benefit from aggressive regularization. Conventional regularization methods, such as dropout or weight decay, do not leverage the structures of the network&#39;s inputs and hidden states. As a result, these conventional methods are less effective than methods that leverage the structures, such as SpatialDropout and DropBlock, which randomly drop the values at certain contiguous areas in the hidden states and setting them to zero. Although the locations of dropping areas random, the patterns of SpatialDropout and DropBlock are manually designed and fixed. Here we propose to learn the dropping patterns. In our method, a controller learns to generate a dropping pattern at every channel and layer of a target network, such as a ConvNet or a Transformer. The target network is then trained with the dropping pattern, and its resulting validation performance is used as a signal for the controller to learn from. We show that this method works well for both image recognition on CIFAR-10 and ImageNet, as well as language modeling on Penn Treebank and WikiText-2. The learned dropping patterns also transfers to different tasks and datasets, such as from language model on Penn Treebank to Engligh-French translation on WMT 2014. Our code will be available at: https://github.com/googleresearch/google-research/tree/master/auto_dropout.

AAAI 2021

AutoDrop: Learning Dropping Patterns to Regularize Deep Networks

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
 The conference will be held in Toronto, Canada, July 9-14, 2023. 
Following the succession of the recent conferences in our field, ACL 2023 will adopt a hybrid format.
While the impact of Covid has considerably diminished in terms of traveling, obtaining visas to Canada
entails a very long process. Moreover, the global economic conditions pose challenges for many individuals to travel to conferences. Recognizing these circumstances, we know many participants may not be
able to attend the conference in person. Therefore, we are committed to providing a great virtual platform
so everyone has the opportunity to interact with other participants and enjoy the conference. Based on the
current registered participants, approxiately 30% have chosen to attend the conference virtually. Whether
you join us in person or virtually, we sincerely hope everyone has a remarkable conference experience. 
This General Chair’s message is where I express my gratitude to the many individuals who have made
enormous contributions to the conference over the past year.

Read [**ACL 2023 General Chair's message**](https://docs.google.com/document/d/1WobYM7norbG4dI48s75HfJoD89qgX5a_F-6U8AteLSA/edit?usp=sharing/) in full.

##### **[Conference Handbook](https://2023.aclweb.org/downloads/acl2023-handbook.pdf)**

ACL 2023

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing.

Neural networks are often over-parameterized and hence benefit from aggressive regularization. Conventional regularization methods, such as dropout or weight decay, do not leverage the structures of the network's inputs and hidden states. As a result, these conventional methods are less effective than methods that leverage the structures, such as SpatialDropout and DropBlock, which randomly drop the values at certain contiguous areas in the hidden states and setting them to zero. Although the locations of dropping areas random, the patterns of SpatialDropout and DropBlock are manually designed and fixed. Here we propose to learn the dropping patterns. In our method, a controller learns to generate a dropping pattern at every channel and layer of a target network, such as a ConvNet or a Transformer. The target network is then trained with the dropping pattern, and its resulting validation performance is used as a signal for the controller to learn from. We show that this method works well for both image recognition on CIFAR-10 and ImageNet, as well as language modeling on Penn Treebank and WikiText-2. The learned dropping patterns also transfers to different tasks and datasets, such as from language model on Penn Treebank to Engligh-French translation on WMT 2014. Our code will be available at: https://github.com/googleresearch/google-research/tree/master/auto_dropout.

technical paper

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

Stochastic gradient descent (SGD) is a widely used method for its outstanding generalization ability and simplicity. Adaptive gradient methods have been proposed to further accelerate the optimization process. In this paper, we revisit existing adaptive gradient optimization methods with a new interpretation. Such new perspective leads to a refreshed understanding of the roles of second moments in stochastic optimization. Based on this, we propose Angle-Calibration Moment method (ACMo), a novel stochastic optimization method. It enjoys the benefits of second moments with only first moment updates. Theoretical analysis shows that ACMo is able to achieve the same convergence rate as mainstream adaptive methods. Experiments on a variety of CV and NLP tasks demonstrate that ACMo has a comparable convergence to state-of-the-art Adam-type optimizers, and even a better generalization performance in most cases. The code is available at https://github.com/Xunpeng746/ACMo.

ACMo: Angle-Calibrated Moment Methods for Stochastic Optimization

In this paper, we study fundamental problems of maximizing DR-submodular continuous functions that have real-world applications in the domain of machine learning, economics, operations research and communication systems. It captures a subclass of non-convex optimization that provides both theoretical and practical guarantees. Here, we focus on minimizing regret for online arriving non-monotone DR-submodular functions over down-closed and general convex sets.
First, we present an online algorithm that achieves a 1/e-approximation ratio with the regret of O(T^{3/4}) for maximizing DR-submodular functions over any down-closed convex set. Note that, the approximation ratio of 1/e matches the best-known guarantee for the offline version of the problem. Next, we give an online algorithm that achieves an approximation guarantee (depending on the search space) for the problem of maximizing non-monotone continuous DR-submodular functions over a general convex set (not necessarily down-closed). To best of our knowledge, no prior algorithm with approximation guarantee was known for non-monotone DR-submodular maximization in the online setting. Finally we run experiments to verify the performance of our algorithms on problems arising in machine learning domain with the real-world datasets.

Online Non-Monotone DR-Submodular Maximization

In this paper, we study a novel problem, Minimum Robust Multi-Submodular Cover for Fairness (MinRF), as follows: given a ground set V; m monotone submodular functions f_1,...,f_m; m thresholds T_1,...,T_m and a non-negative integer r; MinRF asks for the smallest set S such that f_i(S \ X) â‰¥ T_i for all i âˆˆ [m] and |X| â‰¤ r. We prove that MinRF is inapproximable within (1- Îµ) ln m; and no algorithm, taking fewer than exponential number of queries in term of r, is able to output a feasible set to MinRF with high certainty. Three bicriteria approximation algorithms with performance guarantees are proposed: one for r = 0, one for r = 1, and one for general r. We further investigate our algorithms' performance in two applications of MinRF, Information Propagation for Multiple Groups and Movie Recommendation for Multiple Users. Our algorithms have shown to outperform baseline heuristics in both solution quality and the number of queries in most cases.

Minimum Robust Multi-Submodular Cover for Fairness

Slot filling, a fundamental module of spoken language understanding, often suffers from insufficient quantity and diversity of training data. To remedy this, we propose a novel Cluster-to-Cluster generation framework for Data Augmentation (DA), named C2C-GenDA. It enlarges the training set by reconstructing existing utterances into alternative expressions while keeping semantic. Different from previous DA works that reconstruct utterances one by one independently, C2C-GenDA jointly encodes multiple existing utterances of the same semantics and simultaneously decodes multiple unseen expressions. Jointly generating multiple new utterances allows to consider the relations between generated instances and encourages diversity. Besides, encoding multiple existing utterances endows C2C with a wider view of existing expressions, helping to reduce generation that duplicates existing data. Experiments on ATIS and Snips datasets show that instances augmented by C2C-GenDA improve slot filling by 7.99 (11.9%â†‘) and 5.76 (13.6%â†‘) F-scores respectively, when there are only hundreds of training utterances. Code: https://github.com/Sanyuan-Chen/C2C-DA.

C2C-GenDA: Cluster-to-Cluster Generation for Data Augmentation of Slot Filling

The great success of Transformer-based models benefits from the powerful multi-head self-attention mechanism, which learns token dependencies and encodes contextual information from the input. Prior work strives to attribute model decisions to individual input features with different saliency measures, but they fail to explain how these input features interact with each other to reach predictions. In this paper, we propose a self-attention attribution method to interpret the information interactions inside Transformer. We take BERT as an example to conduct extensive studies. Firstly, we apply self-attention attribution to identify the important attention heads, while others can be pruned with marginal performance degradation. Furthermore, we extract the most salient dependencies in each layer to construct an attribution tree, which reveals the hierarchical interactions inside Transformer. Finally, we show that the attribution results can be used as adversarial patterns to implement non-targeted attacks towards BERT.

Self-Attention Attribution: Interpreting Information Interactions Inside Transformer

In this paper, we study the few-shot multi-label classification for user intent detection. For multi-label intent detection, state-of-the-art work estimates label-instance relevance scores and uses a threshold to select multiple associated intent labels. To determine appropriate thresholds with only a few examples, we first learn universal thresholding experience on data-rich domains, and then adapt the thresholds to certain few-shot domains with a calibration based on nonparametric learning. For better calculation of label-instance relevance score, we introduce label name embedding as anchor points in representation space, which refines representations of different classes to be well-separated from each other. Experiments on two datasets show that the proposed model significantly outperforms strong baselines in both one-shot and five-shot settings.

Few-Shot Learning for Multi-Label Intent Detection

Deep neural networks are the default choice of learning models for computer vision tasks. Extensive work has been carried out in recent years on explaining deep models for vision tasks such as classification. However, recent work has shown that it is possible for these models to produce substantially different attribution maps even when two very similar images are given to the network, raising serious questions about trustworthiness. To address this issue, we propose a robust attribution training strategy to improve attributional robustness of deep neural networks. Our method carefully analyzes the requirements for attributional robustness and introduces two new regularizers that preserve a model's attribution map during attacks. Our method surpasses state-of-the-art attributional robustness methods by a margin of approximately 3% to 9% in terms of attribution robustness measures on several datasets including MNIST, FMNIST, Flower and GTSRB.

Enhanced Regularizers for Attributional Robustness

Deep reinforcement learning approaches have shown impressive results in a variety of different domains, however, more complex heterogeneous architectures such as world models require the different neural components to be trained separately instead of end-to-end. While a simple genetic algorithm recently showed end-to-end training is possible, it failed to solve a more complex 3D task. This paper presents a method called Deep Innovation Protection (DIP) that addresses the credit assignment problem in training complex heterogenous neural network models end-to-end for such  environments. The main idea behind the approach is to employ multiobjective optimization to temporally reduce the selection pressure on specific components in multi-component network, allowing other components to adapt. We investigate the emergent representations of these evolved networks, which learn to predict properties important for the survival of the agent, without the need for a specific forward-prediction loss.

Deep Innovation Protection: Confronting the Credit Assignment Problem in Training Heterogeneous Neural Architectures

Explainable AI provides insights to users into the why for
model predictions, offering potential for users to better understand
and trust a model, and to recognize and correct AI
predictions that are incorrect. Prior research on human and
explainable AI interactions has focused on measures such as
interpretability, trust, and usability of the explanation. There
are mixed findings whether explainable AI can improve actual
human decision-making and the ability to identify the
problems with the underlying model. Using real datasets, we
compare objective human decision accuracy without AI (control),
with an AI prediction (no explanation), and AI prediction
with explanation. We find providing any kind of AI prediction
tends to improve user decision accuracy, but no conclusive
evidence that explainable AI has a meaningful impact.
Moreover, we observed the strongest predictor for human decision
accuracy was AI accuracy and that users were somewhat
able to detect when the AI was correct vs. incorrect, but
this was not significantly affected by including an explanation.
Our results indicate that, at least in some situations, the
why information provided in explainable AI may not enhance
user decision-making, and further research may be needed to
understand how to integrate explainable AI into real systems.

Does Explainable Artificial Intelligence Improve Human Decision-Making?

We study the classical problem of matching n agents to n objects, where the agents have ranked preferences over the objects. We focus on two popular desiderata from the matching literature: Pareto optimality and rank-maximality. Instead of asking the agents to report their complete preferences, our goal is to learn a desirable matching from partial preferences, specifically a matching that is necessarily Pareto optimal (NPO) or necessarily rank-maximal (NRM) under any completion of the partial preferences. We focus on the top-k model in which agents reveal a prefix of their preference rankings. We design efficient algorithms to check if a given matching is NPO or NRM, and to check whether such a matching exists given top-k partial preferences. We also study online algorithms for eliciting partial preferences adaptively, and prove bounds on their competitive ratio.

Downloads

Next from AAAI 2021

ACMo: Angle-Calibrated Moment Methods for Stochastic Optimization

Similar lecture

Revisiting Token Dropping Strategy in Efficient BERT Pretraining

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2021

ACMo: Angle-Calibrated Moment Methods for Stochastic Optimization

Similar lecture

Revisiting Token Dropping Strategy in Efficient BERT Pretraining

Downloads