United States

Collaborative learning has successfully applied knowledge transfer to guide a pool of small student networks towards robust local minima. However, previous approaches typically struggle with drastically aggravated student homogenization when the number of students rises. In this paper, we propose Collaborative Group Learning, an efficient framework that aims to diversify the feature representation and conduct an effective regularization. Intuitively, similar to the human group study mechanism, we induce students to learn and exchange different parts of course knowledge as collaborative groups. First, each student is established by randomly routing on a modular neural network, which facilitates flexible knowledge communication between students due to random levels of representation sharing and branching. Second, to resist the student homogenization, students first compose diverse feature sets by exploiting the inductive bias from sub-sets of training data, and then aggregate and distill different complementary knowledge by imitating a random sub-group of students at each time step. Overall, the above mechanisms are beneficial for maximizing the student population to further improve the model generalization without sacrificing computational efficiency. Empirical evaluations on both image and text tasks indicate that our method significantly outperforms various state-of-the-art collaborative approaches whilst enhancing computational efficiency.

AAAI 2021

Collaborative Group Learning

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. 

technical paper

Cognitive psychology research shows that humans have the instinct for abstract thinking, where association plays an essential role in language comprehension. Especially for Chinese, its ideographic writing system allows radicals to trigger semantic association without the need of phonetics. In fact, subconsciously using the associative information guided by radicals is a key for readers to ensure the robustness of semantic understanding. Fortunately, many basic and extended concepts related to radicals are systematically included in Chinese language dictionaries, which leaves a handy but unexplored way for improving Chinese text representation and classification. To this end, we draw inspirations from cognitive principles between ideography and human associative behavior to propose a novel Radical-guided Associative Model (RAM) for Chinese text classification. RAM comprises two coupled spaces, namely Literal Space and Associative Space, which imitates the real process in people's mind when understanding a Chinese text. To be specific, we first devise a serialized modeling structure in Literal Space to thoroughly capture the sequential information of Chinese text. Then, based on the authoritative information provided by Chinese language dictionaries, we design an association module and put forward a strategy called Radical-Word Association to use ideographic radicals as the medium to associate prior concept words in Associative Space. Afterwards, we design an attention module to imitate people's matching and decision between Literal Space and Associative Space, which can balance the importance of each associative words under specific contexts. Finally, extensive experiments on two real-world datasets prove the effectiveness and rationality of RAM, with good cognitive insights for future language modeling.

Ideography Leads Us to the Field of Cognition: A Radical-Guided Associative Model for Chinese Text Classification

Recently, multi-hop reasoning over incomplete Knowledge Graphs (KGs) has attracted wide attention due to its desirable interpretability for downstream tasks, such as question answer and knowledge graph completion. Multi-Hop reasoning is a typical sequential decision problem, which can be formulated as a Markov decision process (MDP). Subsequently, some reinforcement learning (RL) based approaches are proposed and proven effective to train an agent for reasoning paths sequentially until reaching the target answer. However, these approaches assume that an entity/relation representation follows a one-point distribution. In fact, different entities and relations may contain different certainties. On the other hand, since REINFORCE used for updating the policy in these approaches is a biased policy gradients method, the agent is prone to be stuck in high reward paths rather than broad reasoning paths, which leads to premature and suboptimal exploitation. In this paper, we consider a Bayesian reinforcement learning paradigm to harness uncertainty into multi-hop reasoning. By incorporating uncertainty into the representation layer, the agent trained by RL has uncertainty in a region of the state space then it should be more efficient in exploring unknown or less known part of the KG. In our approach, we build a Bayesian Q-learning architecture as a state-action value function for estimating the expected long-term reward. As initialized by Gaussian prior or pre-trained prior distribution, the representation layer drives uncertainty that allows regularizing the training. We conducted extensive experiments on multiple KGs. Experimental results show a superior performance than other baselines, especially significant improvements on the automated extracted KG.

GaussianPath: A Bayesian Multi-Hop Reasoning Framework for Knowledge Graph Reasoning

Research on automatically solving mathematical word problems (MWP) has a long history. Most recent works adopt Seq2Seq approach to predict the result equations as a sequence of quantities and operators. Although result equations can be written as a sequence, it is essentially a structure. More precisely, it is a Direct Acyclic Graph (DAG) whose leaf nodes are the quantities, and internal and root nodes are arithmetic or comparison operators. In this paper, we propose a novel Seq2DAG approach to extract the equation set directly as a DAG structure. It is extracted in a bottom-up fashion by aggregating quantities and sub-expressions layer by layer iteratively. The advantages of our approach approach are three-fold: it is intrinsically suitable to solve multivariate problems, it always outputs valid structure, and its computation satisfies commutative law for +, x and =. Experimental results on Math23K and DRAW1K demonstrate that our model outperforms state-of-the-art deep learning methods. We also conduct detailed analysis on the results to show the strengths and limitations of our approach.

A Bottom-Up DAG Structure Extraction Model for Math Word Problems

Despite the near-human performances already achieved on formal texts such as news articles, neural machine translation still has difficulty in dealing with "user-generated" texts that have diverse linguistic phenomena but lack large-scale high-quality parallel corpora. To address this problem, we propose a counterfactual domain adaptation method to better leverage both large-scale source-domain data (formal texts)  and small-scale target-domain data (informal texts). Specifically, by considering effective counterfactual conditions (the concatenations of source-domain texts and the target-domain tag), we construct the counterfactual representations to fill the sparse latent space of the target domain caused by a small amount of data, that is, bridging the gap between the source-domain data and the target-domain data. Experiments on English-to-Chinese and Chinese-to-English translation tasks show that our method outperforms the base model that is trained only on the informal corpus by a large margin, and consistently surpasses different baseline methods by +1.12 ~ 4.34 BLEU points on different datasets. Furthermore, we also show that our method achieves competitive performances on cross-domain language translation on four language pairs.

Bridging the Domain Gap: Improve Informal Language Translation via Counterfactual Domain Adaptation

Time series shapelets are short discriminative subsequences that recently have been found not only to be accurate but also interpretable for the classification problem of univariate time series (UTS). However, existing work on shapelets selection cannot be applied to multivariate time series classification (MTSC) since the candidate shapelets of MTSC may come from different variables of different lengths and thus cannot be directly compared. To address this challenge, in this paper, we propose a novel model called ShapeNet, which embeds shapelet candidates of different lengths into a unified space for shapelet selection. The network is trained using cluster-wise triplet loss, which considers the distance between anchor and multiple positive (negative) samples and the distance between positive (negative) samples, which are important for convergence. We compute representative and diversified final shapelets rather than directly using all the embeddings for model building to avoid a large fraction of non-discriminative shapelet candidates. We have conducted experiments on ShapeNet with competitive state-of-the-art and benchmark methods using UEA MTS datasets. The results show that the accuracy of ShapeNet is the best of all the methods compared. Furthermore, we illustrate the shapeletsâ€™ interpretability with two case studies.

ShapeNet: A Shapelet-Neural Network Approach for Multivariate Time Series Classification

Estimating the complete 3D point cloud from an incomplete one lies at the core of many vision and robotics applications. Existing methods typically predict the complete point cloud based on the global shape representation extracted from the incomplete input. Although they could predict the overall shape of 3D objects, they are incapable of generating structure details of objects. Moreover, the partial input point sets obtained from range scans are often sparse, noisy and non-uniform, which largely hinder shape completion. In this paper, we propose an adaptive sampling and hierarchical folding network (ASHF-Net) for robust 3D point cloud completion. Our main contributions are two-fold. First, we propose a denoising auto-encoder with an adaptive sampling module, aiming at learning robust local region features that are insensitive to noise. Second, we propose a hierarchical folding decoder with the gated skip-attention and multi-resolution completion goal to effectively exploit the local structure details of partial inputs. We also design a KL regularization term to evenly distribute the generated points. Extensive experiments demonstrate that our method outperforms existing state-of-the-art methods on multiple 3D point cloud completion benchmarks.

ASHF-Net: Adaptive Sampling and Hierarchical Folding Network for Robust Point Cloud Completion

Indoor action recognition plays an important role in modern society, such as intelligent healthcare in large mobile cabin hospitals. With the wide usage of depth sensors like Kinect, multimodal information including skeleton and RGB modalities brings a promising way to improve the performance. However, existing methods are either focusing on a single data modality or failed to take the advantage of multiple data modalities. In this paper, we propose a Teacher-Student Multimodal Fusion (TSMF) model that fuses the skeleton and RGB modalities at the model level for indoor action recognition. In our TSMF, we utilize a teacher network to transfer the structural knowledge of the skeleton modality to a student network for the RGB modality. With extensive experiments on two benchmarking datasets: NTU RGB+D and PKU-MMD, results show that the proposed TSMF consistently performs better than state-of-the-art single modal and multimodal methods. It also indicates that our TSMF could not only improve the accuracy of the student network but also significantly improve the ensemble accuracy.

Multimodal Fusion via Teacher-Student Network for Indoor Action Recognition

Human-object interaction (HOI) detection requires a large amount of annotated data. Current algorithms suffer from insufficient training samples and category imbalance within datasets. To increase data efficiency, in this paper, we propose an efficient and effective data augmentation method called DecAug for HOI detection. Based on our proposed object state similarity metric, object patterns across different HOIs are shared to augment local object appearance features without changing their states. Further, we shift spatial correlation between humans and objects to other feasible configurations with the aid of a pose-guided Gaussian Mixture Model while preserving their interactions. Experiments show that our method brings up to 3.3 mAP and 1.6 mAP improvements on V-COCO and HICO-DET dataset for two advanced models. Specifically, interactions with fewer samples enjoy more notable improvement. Our method can be easily integrated into various HOI detection models with negligible extra computational consumption.

Decaug: Augmenting HOI Detection via Decomposition

Deep reinforcement learning has been successful in a variety of tasks, such as game playing and robotic manipulation. However, attempting to learn tabula rasa disregards the logical structure of many domains as well as the wealth of readily available knowledge from domain experts that could help "warm start" the learning process. We present a novel reinforcement learning technique that allows for intelligent initialization of a neural network weights and architecture. Our approach permits the encoding domain knowledge directly into a neural decision tree, and improves upon that knowledge with policy gradient updates. We empirically validate our approach on two OpenAI Gym tasks and two modified StarCraft 2 tasks, showing that our novel architecture outperforms multilayer-perceptron and recurrent architectures. Our knowledge-based framework finds superior policies compared to imitation learning-based and prior knowledge-based approaches. Importantly, we demonstrate that our approach can be used by untrained humans to initially provide >80% increase in expected reward relative to baselines prior to training (p < 0.001), which results in a >60% increase in expected reward after policy optimization (p = 0.011).

Encoding Human Domain Knowledge to Warm Start Reinforcement Learning

Certifying the output of tools solving complex problems so as to ensure
the correctness of the results they provide is of tremendous importance.
Despite being widespread for SAT-solvers, this level of exigence has
not yet percolated for tools solving more complex tasks, such as model
counting or knowledge compilation. In this paper, the focus is laid on a
general family of top-down Decision-DNNF compilers. We explain how those
compilers can be tweaked so as to output certifiable Decision-DNNF
circuits, which are mainly standard Decision-DNNF circuits decorated by
annotations serving as certificates. We describe a polynomial-time
checker for testing whether a given CNFÂ formula is equivalent or not to
a given certifiable Decision-DNNF circuit. Finally, leveraging a
modified version of the compiler d4 for generating certifiable
Decision-DNNF circuits and an implementation of the checker, we present
the results of an empirical evaluation that has been conducted for
assessing how large are in practice certifiable Decision-DNNF circuits,
and how much time is needed to compute and to check such circuits.

Downloads

Next from AAAI 2021

Ideography Leads Us to the Field of Cognition: A Radical-Guided Associative Model for Chinese Text Classification

Similar lecture

Multi-View Feature Representation for Dialogue Generation with Bidirectional Distillation

.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

Ideography Leads Us to the Field of Cognition: A Radical-Guided Associative Model for Chinese Text Classification

Similar lecture

Multi-View Feature Representation for Dialogue Generation with Bidirectional Distillation

Downloads