United States

Bayesian neural networks (BNNs) retain NN structures with a probability distribution placed over their weights. With the introduced uncertainties and redundancies, BNNs are proper choices of robust controllers for safety-critical control systems. This paper considers the problem of verifying the safety of nonlinear closed-loop systems with BNN controllers over unbounded-time horizon. In essence, we compute a safe weight set such that as long as the BNN controller is always applied with weights sampled from the safe weight set, the controlled system is guaranteed to be safe. We propose a novel two-phase method for the safe weight set computation. First, we construct a reference safe control set that constraints the control inputs, through polynomial approximation to the BNN controller followed by polynomial-optimization-based barrier certificate generation. Then, the computation of safe weight set is reduced to a range inclusion problem of the BNN on the system domain w.r.t. the safe control set, which can be solved incrementally and the set of safe weights can be extracted. Compared with the existing method based on invariant learning and mixed-integer linear programming, we could compute safe weight sets with larger radii on a series of linear benchmarks. Moreover, experiments on a series of widely used nonlinear control tasks show that our method can synthesize large safe weight sets with probability measure as high as 95\% even for a large-scale system of dimension 7.

AAAI 2023

Safety Verification of Nonlinear Systems with Bayesian Neural Network Controllers

safe and robust ai systems

safe learning and control

formal methods for ai systems

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/registration/).


The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines.

poster

Image copy detection (ICD) aims to determine whether a query image is an edited copy of any image from a reference set. Currently, there are very limited public benchmarks for ICD, while all overlook a critical challenge in real-world applications, i.e., the distraction from hard negative queries. Specifically, some queries are not edited copies but are inherently similar to some reference images. These hard negative queries are easily false recognized as edited copies, significantly compromising the ICD accuracy. This observation motivates us to build the first ICD benchmark featuring this characteristic. Based on existing ICD datasets, this paper constructs a new dataset by additionally adding 100,000 and 24, 252 hard negative pairs into the training and test set, respectively. Moreover, this paper further reveals a unique difficulty for solving the hard negative problem in ICD, i.e., there is a fundamental conflict between current metric learning and ICD. This conflict is: the metric learning adopts symmetric distance while the edited copy is an asymmetric (unidirectional) process, e.g., a partial crop is close to its holistic reference image and is an edited copy, while the latter can- not be the edited copy of the former (in spite the distance is equally small). This insight results in an Asymmetrical- Similarity Learning (ASL) method, which allows the similarity in two directions (the query ↔ the reference image) to be different from each other. Experimental results show that ASL outperforms state-of-the-art methods by a clear margin, confirming that solving the symmetric-asymmetric conflict is critical for ICD. The NDEC dataset and code are available at https://github.com/WangWenhao0716/ASL.

A Benchmark and Asymmetrical-Similarity Learning for Practical Image Copy Detection

The increased integration of artificial intelligence (AI) technologies in human workflows has resulted in a new paradigm of AI-assisted decision making,
in which an AI model provides decision recommendations while humans make the final decisions. To best support humans in decision making, it is critical to obtain a quantitative understanding of how humans interact with and rely on AI. Previous studies often model humans' reliance on AI as an analytical process, i.e., reliance decisions are made based on cost-benefit analysis. However, theoretical models in psychology suggest that the reliance decisions can often be driven by {\em emotions} like humans' {\em trust} in AI models. In this paper, we propose a {\em hidden Markov model} to capture the affective process underlying the human-AI interaction in AI-assisted decision making, by characterizing how decision makers adjust their trust in AI over time and make reliance decisions based on their trust. Evaluations on real human behavior data collected from human-subject experiments show that the proposed model outperforms various baselines in accurately predicting humans' reliance behavior in AI-assisted decision making. Based on the proposed model, we further provide insights into how humans' trust and reliance dynamics in AI-assisted decision making is influenced by contextual factors like decision stakes and their interaction experiences. 

Modeling Human Trust and Reliance in AI-assisted Decision Making: A Markovian Approach

Different people speak with diverse personalized speaking styles. Although existing one-shot talking head methods have made significant progress in lip sync, natural facial expressions, and stable head motions, they still cannot generate diverse speaking styles in the final talking head videos. To tackle this problem, we propose a one-shot style-controllable talking face generation framework. In a nutshell, we aim to attain a speaking style from an arbitrary reference speaking video and then drive the one-shot portrait to speak with the reference speaking style and another piece of audio. Specifically, we first develop a style encoder to extract dynamic facial motion patterns of a style reference video and then encode them into a style code. Afterward, we introduce a style-controllable decoder to synthesize stylized facial animations from the speech content and style code. In order to integrate the reference speaking style into generated videos, we design a style-aware adaptive transformer, which enables the encoded style code to adjust the weights of the feed-forward layers accordingly. Thanks to the style-aware adaptation mechanism, the reference speaking style can be better embedded into synthesized videos during decoding. Extensive experiments demonstrate that our method is capable of generating talking head videos with diverse speaking styles from only one portrait image and an audio clip while achieving authentic visual effects.

StyleTalk: One-shot Talking Head Generation with Controllable Speaking Styles

Weisfeiler-Lehman (WL) test is widely applied as a basic algorithm for graph machine learning models, such as graph kernels, graph metrics, and graph neural networks. However, its ability to describe structural information is insufficient because it focuses only on the consistency of graphs. Therefore, this limits the performance of existing methods based on the WL test. For example, the similarity and distance between graphs defined using the WL test are coarse measures that do not accurately describe the differences in structure. This paper clarifies these facts and proposes a novel graph metric called Wasserstein WL subtree (WWLS) distance. First, we adopt the WL subtree as the structural information of the node neighborhood. Then, we define the node metric using L1-approximated tree edit distance (L1-TED) between WL subtrees of nodes. Finally, we define the WWLS by combining the Wasserstein distance and the L1-TED. The WWLS can capture slight structural differences that are difficult with conventional metrics. We demonstrate its superior performance in metric validation and graph classification experiments.

Wasserstein Graph Distance based on L1‚ÄìApproximated Tree Edit Distance between Weisfeiler‚ÄìLehman Subtrees

This paper presents a rewriting method for Boolean circuits that minimizes small subcircuits with exact synthesis. Individual synthesis tasks are encoded as Quantified Boolean Formulas (QBFs) that capture the full flexibility for the implementation of multi-output subcircuits.
This is in contrast to SAT-based resynthesis, where "don't cares" are computed for an individual gate, and replacements are confined to circuitry that is used exclusively by that gate.
An implementation of our method achieved substantial size reductions compared to state-of-the-art methods across a wide range of benchmark circuits.

Circuit Minimization with QBF-Based Exact Synthesis

Achieving gender equality is an important pillar for humankind’s sustainable future. Pioneering data-driven gender bias research is based on large-scale public records such as scientific papers, patents, and company registrations, covering female researchers, inventors and entrepreneurs, and so on. Since gender information is often missing in relevant datasets, studies rely on tools to infer genders from names. However, available open-sourced Chinese gender-guessing tools are not yet suitable for scientific purposes, which may be partially responsible for female Chinese being underrepresented in mainstream gender bias research and affect their universality. Specifically, these tools focus on character-level information while overlooking the fact that the combinations of Chinese characters in multi-character names, as well as the components and pronunciations of characters, convey important messages. As a first effort, we design a Chinese Heterogeneous Graph Attention (CHGAT) model to capture the heterogeneity in component relationships and incorporate the pronunciations of characters. Our model largely surpasses current tools and also outperforms the state-of-the-art algorithm. Last but not least, the most popular Chinese name-gender dataset is single-character based with far less female coverage from an unreliable source, naturally hindering relevant studies. We open-source a more balanced multi-character dataset from an official source together with our code, hoping to help future research promoting gender equality.

For the Underrepresented in Gender Bias Research: Chinese Name Gender Prediction with Heterogeneous Graph Attention Network

Trustworthy Residual Vehicle Value Prediction for Auto Finance

Deep learning models have achieved tremendous successes in accurate predictions for computer vision, natural language processing and speech recognition applications. However, to succeed in high-risk and safety-critical domains such as healthcare and finance, these deep learning models need to be made reliable and trustworthy. Specifically, they need to be robust and adaptive to real-world environments which can be drastically different from the training settings. In this talk, I will advocate for Bayesian principles to achieve the goal of building robust and adaptive deep learning models. I will introduce a suite of uncertainty quantification methods for Bayesian deep learning, and demonstrate applications enabled by accurate uncertainty estimates, e.g., robust prediction, continual learning and repairing model failures. I will conclude by discussing the research challenges for robust and adaptive deep learning models.

Robust and Adaptive Deep Learning via Bayesian Principles

In recent years, disinformation and fake news have become a global phenomenon, particularly so during the pandemic, COVID-19. The wide dissemination of fake news can have detrimental societal effects on individuals and the society. Despite the recent progress in detecting fake news, it remains a challenging task due to its complexity, diversity, multi-modality, speed, and costs of fact-checking or annotation. In this talk, we look at some challenges and lessons learned when exploring effective and explainable detecting strategies of fake news, and discuss challenges in fake news research and the need for interdisciplinary research.

Combating Disinformation on Social Media and Its Challenges: A Computational Perspective

In the talk, I shall describe my lab’s recent advances in AI, applied machine learning, and data mining to combat malicious actors (sockpuppets, ban evaders, etc.) and dangerous content (misinformation, hate, etc.) on web and social media platforms. My vision is to create a trustworthy online ecosystem for everyone and create the next generation of socially-aware methods that promote health, equity, and safety. Broadly, in my research, I have created novel graph, content (NLP, multimodality), and adversarial machine learning methods leveraging terabytes of data to detect, predict, and mitigate online threats. My interdisciplinary research innovates socio-technical solutions that I achieve by amalgamating computer science with social science theories. I am also passionate about putting my re- search into practice – my lab’s models have been deployed on Flipkart, influenced Twitter’s Birdwatch, and now being deployed on Wikipedia. My research has started a paradigm shift from the current slow and reactive approach against online harms to agile, proactive, and whole-of-society solutions. My talk will outline four thrusts of my research:
(1) Detection of harmful content and malicious actors across platforms, languages, and modalities
(2) Robustifying detection models against adversarial actors by predicting future malicious activities
(3) Attributing the impact of harmful content and the role of recommender systems
(4) Mitigation techniques to counter misinformation by professionals and non-expert crowds

Premium content

Safety Verification of Nonlinear Systems with Bayesian Neural Network Controllers

Downloads

Next from AAAI 2023

A Benchmark and Asymmetrical-Similarity Learning for Practical Image Copy Detection

Similar lecture

Towards Safe AI: Sandboxing DNNs-based Controllers in Stochastic Games

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES