arXiv subjects

Yizhou Zhao

Publications and source records attributed to Yizhou Zhao.

At least 55 records · Page 3Linked to original sources

arXiv2023

Weakly Secure Summation with Colluding Users

In secure summation, $K$ users, each holds an input, wish to compute the sum of the inputs at a server without revealing any information about {\em all the inputs} even if the server may collude with {\em an arbitrary subset of users}. In this work, we relax the security and colluding constraints, where the set of inputs whose information is prohibited from leakage is from a predetermined collection of sets (e.g., any set of up to $S$ inputs) and the set of colluding users is from another predetermined collection of sets (e.g., any set of up to $T$ users). For arbitrary collection of security input sets and colluding user sets, we characterize the optimal randomness assumption, i.e., the minimum number of key bits that need to be held by the users, per input bit, for weakly secure summation to be feasible, which generally involves solving a linear program.

cs.IT↗

arXiv2022

Alignment-guided Temporal Attention for Video Action Recognition

Temporal modeling is crucial for various video learning tasks. Most recent approaches employ either factorized (2D+1D) or joint (3D) spatial-temporal operations to extract temporal contexts from the input frames. While the former is more efficient in computation, the latter often obtains better performance. In this paper, we attribute this to a dilemma between the sufficiency and the efficiency of interactions among various positions in different frames. These interactions affect the extraction of task-relevant information shared among frames. To resolve this issue, we prove that frame-by-frame alignments have the potential to increase the mutual information between frame representations, thereby including more task-relevant information to boost effectiveness. Then we propose Alignment-guided Temporal Attention (ATA) to extend 1-dimensional temporal attention with parameter-free patch-level alignments between neighboring frames. It can act as a general plug-in for image backbones to conduct the action recognition task without any model-specific design. Extensive experiments on multiple benchmarks demonstrate the superiority and generality of our module.

cs.CV↗

arXiv2022

OpenD: A Benchmark for Language-Driven Door and Drawer Opening

We introduce OPEND, a benchmark for learning how to use a hand to open cabinet doors or drawers in a photo-realistic and physics-reliable simulation environment driven by language instruction. To solve the task, we propose a multi-step planner composed of a deep neural network and rule-base controllers. The network is utilized to capture spatial relationships from images and understand semantic meaning from language instructions. Controllers efficiently execute the plan based on the spatial and semantic understanding. We evaluate our system by measuring its zero-shot performance in test data set. Experimental results demonstrate the effectiveness of decision planning by our multi-step planner for different hands, while suggesting that there is significant room for developing better models to address the challenge brought by language understanding, spatial reasoning, and long-term manipulation. We will release OPEND and host challenges to promote future research in this area.

cs.LG↗

arXiv2022

MDS Variable Generation and Secure Summation with User Selection

A collection of $K$ random variables are called $(K,n)$-MDS if any $n$ of the $K$ variables are independent and determine all remaining variables. In the MDS variable generation problem, $K$ users wish to generate variables that are $(K,n)$-MDS using a randomness variable owned by each user. We show that to generate $1$ bit of $(K,n)$-MDS variables for each $n \in \{1,2,\cdots, K\}$, the minimum size of the randomness variable at each user is $1 + 1/2 + \cdots + 1/K$ bits. An intimately related problem is secure summation with user selection, where a server may select an arbitrary subset of $K$ users and securely compute the sum of the inputs of the selected users. We show that to compute $1$ bit of an arbitrarily chosen sum securely, the minimum size of the key held by each user is $1 + 1/2 + \cdots + 1/(K-1)$ bits, whose achievability uses the generation of $(K,n)$-MDS variables for $n \in \{1,2,\cdots,K-1\}$.

cs.IT↗

arXiv2022

Learning to Act with Affordance-Aware Multimodal Neural SLAM

Recent years have witnessed an emerging paradigm shift toward embodied artificial intelligence, in which an agent must learn to solve challenging tasks by interacting with its environment. There are several challenges in solving embodied multimodal tasks, including long-horizon planning, vision-and-language grounding, and efficient exploration. We focus on a critical bottleneck, namely the performance of planning and navigation. To tackle this challenge, we propose a Neural SLAM approach that, for the first time, utilizes several modalities for exploration, predicts an affordance-aware semantic map, and plans over it at the same time. This significantly improves exploration efficiency, leads to robust long-horizon planning, and enables effective vision-and-language grounding. With the proposed Affordance-aware Multimodal Neural SLAM (AMSLAM) approach, we obtain more than 40% improvement over prior published work on the ALFRED benchmark and set a new state-of-the-art generalization performance at a success rate of 23.48% on the test unseen scenes.

cs.RO↗

arXiv2022

IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language Reasoning

Current visual question answering (VQA) tasks mainly consider answering human-annotated questions for natural images. However, aside from natural images, abstract diagrams with semantic richness are still understudied in visual understanding and reasoning research. In this work, we introduce a new challenge of Icon Question Answering (IconQA) with the goal of answering a question in an icon image context. We release IconQA, a large-scale dataset that consists of 107,439 questions and three sub-tasks: multi-image-choice, multi-text-choice, and filling-in-the-blank. The IconQA dataset is inspired by real-world diagram word problems that highlight the importance of abstract diagram understanding and comprehensive cognitive reasoning. Thus, IconQA requires not only perception skills like object recognition and text understanding, but also diverse cognitive reasoning skills, such as geometric reasoning, commonsense reasoning, and arithmetic reasoning. To facilitate potential IconQA models to learn semantic representations for icon images, we further release an icon dataset Icon645 which contains 645,687 colored icons on 377 classes. We conduct extensive user studies and blind experiments and reproduce a wide range of advanced VQA methods to benchmark the IconQA task. Also, we develop a strong IconQA baseline Patch-TRM that applies a pyramid cross-modal Transformer with input diagram embeddings pre-trained on the icon dataset. IconQA and Icon645 are available at https://iconqa.github.io.

cs.CV↗

arXiv2022

Towards Socially Intelligent Agents with Mental State Transition and Human Utility

Building a socially intelligent agent involves many challenges. One of which is to track the agent's mental state transition and teach the agent to make decisions guided by its value like a human. Towards this end, we propose to incorporate mental state simulation and value modeling into dialogue agents. First, we build a hybrid mental state parser that extracts information from both the dialogue and event observations and maintains a graphical representation of the agent's mind; Meanwhile, the transformer-based value model learns human preferences from the human value dataset, ValueNet. Empirical results show that the proposed model attains state-of-the-art performance on the dialogue/action/emotion prediction task in the fantasy text-adventure game dataset, LIGHT. We also show example cases to demonstrate: (i) how the proposed mental state parser can assist the agent's decision by grounding on the context like locations and objects, and (ii) how the value model can help the agent make decisions based on its personal priorities.

cs.CL↗

arXiv2022

VRKitchen2.0-IndoorKit: A Tutorial for Augmented Indoor Scene Building in Omniverse

With the recent progress of simulations by 3D modeling software and game engines, many researchers have focused on Embodied AI tasks in the virtual environment. However, the research community lacks a platform that can easily serve both indoor scene synthesis and model benchmarking with various algorithms. Meanwhile, computer graphics-related tasks need a toolkit for implementing advanced synthesizing techniques. To facilitate the study of indoor scene building methods and their potential robotics applications, we introduce INDOORKIT: a built-in toolkit for NVIDIA OMNIVERSE that provides flexible pipelines for indoor scene building, scene randomizing, and animation controls. Besides, combining Python coding in the animation software INDOORKIT assists researchers in creating real-time training and controlling avatars and robotics. The source code for this toolkit is available at https://github.com/realvcla/VRKitchen2.0-Tutorial, and the tutorial along with the toolkit is available at https://vrkitchen20-tutorial.readthedocs.io/en/

cs.CG↗

arXiv2022

Secure Summation: Capacity Region, Groupwise Key, and Feasibility

The secure summation problem is considered, where $K$ users, each holds an input, wish to compute the sum of their inputs at a server securely, i.e., without revealing any information beyond the sum even if the server may collude with any set of up to $T$ users. First, we prove a folklore result for secure summation - to compute $1$ bit of the sum securely, each user needs to send at least $1$ bit to the server, each user needs to hold a key of at least $1$ bit, and all users need to hold collectively some key variables of at least $K-1$ bits. Next, we focus on the symmetric groupwise key setting, where every group of $G$ users share an independent key. We show that for symmetric groupwise keys with group size $G$, when $G > K-T$, the secure summation problem is not feasible; when $G \leq K-T$, to compute $1$ bit of the sum securely, each user needs to send at least $1$ bit to the server and the size of each groupwise key is at least $(K-T-1)/\binom{K-T}{G}$ bits. Finally, we relax the symmetry assumption on the groupwise keys and the colluding user sets; we allow any arbitrary group of users to share an independent key and any arbitrary group of users to collude with the server. For such a general groupwise key and colluding user setting, we show that secure summation is feasible if and only if the hypergraph, where each node is a user and each edge is a group of users sharing the same key, is connected after removing the nodes corresponding to any colluding set of users and their incident edges.

cs.IT↗

arXiv2022

Semantic-aligned Fusion Transformer for One-shot Object Detection

One-shot object detection aims at detecting novel objects according to merely one given instance. With extreme data scarcity, current approaches explore various feature fusions to obtain directly transferable meta-knowledge. Yet, their performances are often unsatisfactory. In this paper, we attribute this to inappropriate correlation methods that misalign query-support semantics by overlooking spatial structures and scale variances. Upon analysis, we leverage the attention mechanism and propose a simple but effective architecture named Semantic-aligned Fusion Transformer (SaFT) to resolve these issues. Specifically, we equip SaFT with a vertical fusion module (VFM) for cross-scale semantic enhancement and a horizontal fusion module (HFM) for cross-sample feature fusion. Together, they broaden the vision for each feature point from the support to a whole augmented feature pyramid from the query, facilitating semantic-aligned associations. Extensive experiments on multiple benchmarks demonstrate the superiority of our framework. Without fine-tuning on novel classes, it brings significant performance gains to one-stage baselines, lifting state-of-the-art results to a higher level.

cs.CV↗

arXiv2022

SocAoG: Incremental Graph Parsing for Social Relation Inference in Dialogues

Inferring social relations from dialogues is vital for building emotionally intelligent robots to interpret human language better and act accordingly. We model the social network as an And-or Graph, named SocAoG, for the consistency of relations among a group and leveraging attributes as inference cues. Moreover, we formulate a sequential structure prediction task, and propose an $α$-$β$-$γ$ strategy to incrementally parse SocAoG for the dynamic inference upon any incoming utterance: (i) an $α$ process predicting attributes and relations conditioned on the semantics of dialogues, (ii) a $β$ process updating the social relations based on related attributes, and (iii) a $γ$ process updating individual's attributes based on interpersonal social relations. Empirical results on DialogRE and MovieGraph show that our model infers social relations more accurately than the state-of-the-art methods. Moreover, the ablation study shows the three processes complement each other, and the case study demonstrates the dynamic relational inference.

cs.CL↗

arXiv2022

Triangular Character Animation Sampling with Motion, Emotion, and Relation

Dramatic progress has been made in animating individual characters. However, we still lack automatic control over activities between characters, especially those involving interactions. In this paper, we present a novel energy-based framework to sample and synthesize animations by associating the characters' body motions, facial expressions, and social relations. We propose a Spatial-Temporal And-Or graph (ST-AOG), a stochastic grammar model, to encode the contextual relationship between motion, emotion, and relation, forming a triangle in a conditional random field. We train our model from a labeled dataset of two-character interactions. Experiments demonstrate that our method can recognize the social relation between two characters and sample new scenes of vivid motion and emotion using Markov Chain Monte Carlo (MCMC) given the social relation. Thus, our method can provide animators with an automatic way to generate 3D character animations, help synthesize interactions between Non-Player Characters (NPCs), and enhance machine emotion intelligence (EQ) in virtual reality (VR).

cs.GR↗

arXiv2021

Learning from the Tangram to Solve Mini Visual Tasks

Current pre-training methods in computer vision focus on natural images in the daily-life context. However, abstract diagrams such as icons and symbols are common and important in the real world. This work is inspired by Tangram, a game that requires replicating an abstract pattern from seven dissected shapes. By recording human experience in solving tangram puzzles, we present the Tangram dataset and show that a pre-trained neural model on the Tangram helps solve some mini visual tasks based on low-resolution vision. Extensive experiments demonstrate that our proposed method generates intelligent solutions for aesthetic tasks such as folding clothes and evaluating room layouts. The pre-trained feature extractor can facilitate the convergence of few-shot learning tasks on human handwriting and improve the accuracy in identifying icons by their contours. The Tangram dataset is available at https://github.com/yizhouzhao/Tangram.

cs.CV↗

arXiv2021

ValueNet: A New Dataset for Human Value Driven Dialogue System

Building a socially intelligent agent involves many challenges, one of which is to teach the agent to speak guided by its value like a human. However, value-driven chatbots are still understudied in the area of dialogue systems. Most existing datasets focus on commonsense reasoning or social norm modeling. In this work, we present a new large-scale human value dataset called ValueNet, which contains human attitudes on 21,374 text scenarios. The dataset is organized in ten dimensions that conform to the basic human value theory in intercultural research. We further develop a Transformer-based value regression model on ValueNet to learn the utility distribution. Comprehensive empirical results show that the learned value model could benefit a wide range of dialogue tasks. For example, by teaching a generative agent with reinforcement learning and the rewards from the value model, our method attains state-of-the-art performance on the personalized dialog generation dataset: Persona-Chat. With values as additional features, existing emotion recognition models enable capturing rich human emotions in the context, which further improves the empathetic response generation performance in the EmpatheticDialogues dataset. To the best of our knowledge, ValueNet is the first large-scale text dataset for human value modeling, and we are the first one trying to incorporate a value model into emotionally intelligent dialogue systems. The dataset is available at https://liang-qiu.github.io/ValueNet/.

cs.CL↗

arXiv2021

GenMotion: Data-driven Motion Generators for Real-time Animation Synthesis

With the recent success of deep learning algorithms, many researchers have focused on generative models for human motion animation. However, the research community lacks a platform for training and benchmarking various algorithms, and the animation industry needs a toolkit for implementing advanced motion synthesizing techniques. To facilitate the study of deep motion synthesis methods for skeleton-based human animation and their potential applications in practical animation making, we introduce \genmotion: a library that provides unified pipelines for data loading, model training, and animation sampling with various deep learning algorithms. Besides, by combining Python coding in the animation software \genmotion\ can assist animators in creating real-time 3D character animation. Source code is available at https://github.com/realvcla/GenMotion/.

cs.GR↗

arXiv2021

LUMINOUS: Indoor Scene Generation for Embodied AI Challenges

Learning-based methods for training embodied agents typically require a large number of high-quality scenes that contain realistic layouts and support meaningful interactions. However, current simulators for Embodied AI (EAI) challenges only provide simulated indoor scenes with a limited number of layouts. This paper presents Luminous, the first research framework that employs state-of-the-art indoor scene synthesis algorithms to generate large-scale simulated scenes for Embodied AI challenges. Further, we automatically and quantitatively evaluate the quality of generated indoor scenes via their ability to support complex household tasks. Luminous incorporates a novel scene generation algorithm (Constrained Stochastic Scene Generation (CSSG)), which achieves competitive performance with human-designed scenes. Within Luminous, the EAI task executor, task instruction generation module, and video rendering toolkit can collectively generate a massive multimodal dataset of new scenes for the training and evaluation of Embodied AI agents. Extensive experimental results demonstrate the effectiveness of the data generated by Luminous, enabling the comprehensive assessment of embodied agents on generalization and robustness.

cs.AI↗

arXiv2021

STAR: Sparse Transformer-based Action Recognition

The cognitive system for human action and behavior has evolved into a deep learning regime, and especially the advent of Graph Convolution Networks has transformed the field in recent years. However, previous works have mainly focused on over-parameterized and complex models based on dense graph convolution networks, resulting in low efficiency in training and inference. Meanwhile, the Transformer architecture-based model has not yet been well explored for cognitive application in human action and behavior estimation. This work proposes a novel skeleton-based human action recognition model with sparse attention on the spatial dimension and segmented linear attention on the temporal dimension of data. Our model can also process the variable length of video clips grouped as a single batch. Experiments show that our model can achieve comparable performance while utilizing much less trainable parameters and achieve high speed in training and inference. Experiments show that our model achieves 4~18x speedup and 1/7~1/15 model size compared with the baseline models at competitive accuracy.

cs.CV↗

arXiv2021

Information Theoretic Secure Aggregation with User Dropouts

In the robust secure aggregation problem, a server wishes to learn and only learn the sum of the inputs of a number of users while some users may drop out (i.e., may not respond). The identity of the dropped users is not known a priori and the server needs to securely recover the sum of the remaining surviving users. We consider the following minimal two-round model of secure aggregation. Over the first round, any set of no fewer than $U$ users out of $K$ users respond to the server and the server wants to learn the sum of the inputs of all responding users. The remaining users are viewed as dropped. Over the second round, any set of no fewer than $U$ users of the surviving users respond (i.e., dropouts are still possible over the second round) and from the information obtained from the surviving users over the two rounds, the server can decode the desired sum. The security constraint is that even if the server colludes with any $T$ users and the messages from the dropped users are received by the server (e.g., delayed packets), the server is not able to infer any additional information beyond the sum in the information theoretic sense. For this information theoretic secure aggregation problem, we characterize the optimal communication cost. When $U \leq T$, secure aggregation is not feasible, and when $U > T$, to securely compute one symbol of the sum, the minimum number of symbols sent from each user to the server is $1$ over the first round, and $1/(U-T)$ over the second round.

cs.IT↗