arXiv subjects

Tianyu Wang

Publications and source records attributed to Tianyu Wang.

At least 181 records · Page 10Linked to original sources

arXiv2023

Transfer Learning for Hate Speech Detection in Social Media

Today, the internet is an integral part of our daily lives, enabling people to be more connected than ever before. However, this greater connectivity and access to information increase exposure to harmful content such as cyber-bullying and cyber-hatred. Models based on machine learning and natural language offer a way to make online platforms safer by identifying hate speech in web text autonomously. However, the main difficulty is annotating a sufficiently large number of examples to train these models. This paper uses a transfer learning technique to leverage two independent datasets jointly and builds a single representation of hate speech. We build an interpretable two-dimensional visualization tool of the constructed hate speech representation -- dubbed the Map of Hate -- in which multiple datasets can be projected and comparatively analyzed. The hateful content is annotated differently across the two datasets (racist and sexist in one dataset, hateful and offensive in another). However, the common representation successfully projects the harmless class of both datasets into the same space and can be used to uncover labeling errors (false positives). We also show that the joint representation boosts prediction performances when only a limited amount of supervision is available. These methods and insights hold the potential for safer social media and reduce the need to expose human moderators and annotators to distressing online messaging.

cs.SI↗

arXiv2023

Linear RNNs Provably Learn Linear Dynamic Systems

We study the learning ability of linear recurrent neural networks with Gradient Descent. We prove the first theoretical guarantee on linear RNNs to learn any stable linear dynamic system using any a large type of loss functions. For an arbitrary stable linear system with a parameter $ρ_C$ related to the transition matrix $C$, we show that despite the non-convexity of the parameter optimization loss if the width of the RNN is large enough (and the required width in hidden layers does not rely on the length of the input sequence), a linear RNN can provably learn any stable linear dynamic system with the sample and time complexity polynomial in $\frac{1}{1-ρ_C}$. Our results provide the first theoretical guarantee to learn a linear RNN and demonstrate how can the recurrent structure help to learn a dynamic system.

cs.LG↗

arXiv2023

The hardware is the software

Human brains and bodies are not hardware running software: the hardware is the software. We reason that because the microscopic physics of artificial-intelligence hardware and of human biological "hardware" is distinct, neuromorphic engineers need to be cautious (and yet also creative) in how we take inspiration from biological intelligence. We should focus primarily on principles and design ideas that respect -- and embrace -- the underlying hardware physics of non-biological intelligent systems, rather than abstracting it away. We see a major role for neuroscience in neuromorphic computing as identifying the physics-agnostic principles of biological intelligence -- that is the principles of biological intelligence that can be gainfully adapted and applied to any physical hardware.

cs.ET↗

arXiv2023

Variational Inference for Scalable 3D Object-centric Learning

We tackle the task of scalable unsupervised object-centric representation learning on 3D scenes. Existing approaches to object-centric representation learning show limitations in generalizing to larger scenes as their learning processes rely on a fixed global coordinate system. In contrast, we propose to learn view-invariant 3D object representations in localized object coordinate systems. To this end, we estimate the object pose and appearance representation separately and explicitly map object representations across views while maintaining object identities. We adopt an amortized variational inference pipeline that can process sequential input and scalably update object latent distributions online. To handle large-scale scenes with a varying number of objects, we further introduce a Cognitive Map that allows the registration and query of objects on a per-scene global map to achieve scalable representation learning. We explore the object-centric neural radiance field (NeRF) as our 3D scene representation, which is jointly modeled within our unsupervised object-centric learning framework. Experimental results on synthetic and real datasets show that our proposed method can infer and maintain object-centric representations of 3D scenes and outperforms previous models.

cs.CV↗

arXiv2023

Hedging Complexity in Generalization via a Parametric Distributionally Robust Optimization Framework

Empirical risk minimization (ERM) and distributionally robust optimization (DRO) are popular approaches for solving stochastic optimization problems that appear in operations management and machine learning. Existing generalization error bounds for these methods depend on either the complexity of the cost function or dimension of the random perturbations. Consequently, the performance of these methods can be poor for high-dimensional problems with complex objective functions. We propose a simple approach in which the distribution of random perturbations is approximated using a parametric family of distributions. This mitigates both sources of complexity; however, it introduces a model misspecification error. We show that this new source of error can be controlled by suitable DRO formulations. Our proposed parametric DRO approach has significantly improved generalization bounds over existing ERM and DRO methods and parametric ERM for a wide variety of settings. Our method is particularly effective under distribution shifts and works broadly in contextual optimization. We also illustrate the superior performance of our approach on both synthetic and real-data portfolio optimization and regression tasks.

math.OC↗

arXiv2023

Anisotropic body compliance facilitates robotic sidewinding in complex environments

Sidewinding, a locomotion strategy characterized by the coordination of lateral and vertical body undulations, is frequently observed in rattlesnakes and has been successfully reconstructed by limbless robotic systems for effective movement across diverse terrestrial terrains. However, the integration of compliant mechanisms into sidewinding limbless robots remains less explored, posing challenges for navigation in complex, rheologically diverse environments. Inspired by a notable control simplification via mechanical intelligence in lateral undulation, which offloads feedback control to passive body mechanics and interactions with the environment, we present an innovative design of a mechanically intelligent limbless robot for sidewinding. This robot features a decentralized bilateral cable actuation system that resembles organismal muscle actuation mechanisms. We develop a feedforward controller that incorporates programmable body compliance into the sidewinding gait template. Our experimental results highlight the emergence of mechanical intelligence when the robot is equipped with an appropriate level of body compliance. This allows the robot to 1) locomote more energetically efficiently, as evidenced by a reduced cost of transport, and 2) navigate through terrain heterogeneities, all achieved in an open-loop manner, without the need for environmental awareness.

cs.RO↗

arXiv2023

Robust self-propulsion in sand using simply controlled vibrating cubes

Much of the Earth and many surfaces of extraterrestrial bodies are composed of in-cohesive particle matter. Locomoting on granular terrain is challenging for common robotic devices, either wheeled or legged. In this work, we discover a robust alternative locomotion mechanism on granular media -- generating movement via self-vibration. To demonstrate the effectiveness of this locomotion mechanism, we develop a cube-shaped robot with an embedded vibratory motor and conduct systematic experiments on diverse granular terrains of various particle properties. We investigate how locomotion changes as a function of vibration frequency/intensity on granular terrains. Compared to hard surfaces, we find such a vibratory locomotion mechanism enables the robot to move faster, and more stable on granular surfaces, facilitated by the interaction between the body and surrounding granules. The simplicity in structural design and controls of this robotic system indicates that vibratory locomotion can be a valuable alternative way to produce robust locomotion on granular terrains. We further demonstrate that such cube-shape robots can be used as modular units for morphologically structured vibratory robots with capabilities of maneuverable forward and turning motions, showing potential practical scenarios for robotic systems.

cs.RO↗

arXiv2023

Dynamic Handover: Throw and Catch with Bimanual Hands

Humans throw and catch objects all the time. However, such a seemingly common skill introduces a lot of challenges for robots to achieve: The robots need to operate such dynamic actions at high-speed, collaborate precisely, and interact with diverse objects. In this paper, we design a system with two multi-finger hands attached to robot arms to solve this problem. We train our system using Multi-Agent Reinforcement Learning in simulation and perform Sim2Real transfer to deploy on the real robots. To overcome the Sim2Real gap, we provide multiple novel algorithm designs including learning a trajectory prediction model for the object. Such a model can help the robot catcher has a real-time estimation of where the object will be heading, and then react accordingly. We conduct our experiments with multiple objects in the real-world system, and show significant improvements over multiple baselines. Our project page is available at \url{https://binghao-huang.github.io/dynamic_handover/}.

cs.RO↗

arXiv2023

WALL-E: Embodied Robotic WAiter Load Lifting with Large Language Model

Enabling robots to understand language instructions and react accordingly to visual perception has been a long-standing goal in the robotics research community. Achieving this goal requires cutting-edge advances in natural language processing, computer vision, and robotics engineering. Thus, this paper mainly investigates the potential of integrating the most recent Large Language Models (LLMs) and existing visual grounding and robotic grasping system to enhance the effectiveness of the human-robot interaction. We introduce the WALL-E (Embodied Robotic WAiter load lifting with Large Language model) as an example of this integration. The system utilizes the LLM of ChatGPT to summarize the preference object of the users as a target instruction via the multi-round interactive dialogue. The target instruction is then forwarded to a visual grounding system for object pose and size estimation, following which the robot grasps the object accordingly. We deploy this LLM-empowered system on the physical robot to provide a more user-friendly interface for the instruction-guided grasping task. The further experimental results on various real-world scenarios demonstrated the feasibility and efficacy of our proposed framework. See the project website at: https://star-uu-wang.github.io/WALL-E/

cs.RO↗

arXiv2023

SILT: Shadow-aware Iterative Label Tuning for Learning to Detect Shadows from Noisy Labels

Existing shadow detection datasets often contain missing or mislabeled shadows, which can hinder the performance of deep learning models trained directly on such data. To address this issue, we propose SILT, the Shadow-aware Iterative Label Tuning framework, which explicitly considers noise in shadow labels and trains the deep model in a self-training manner. Specifically, we incorporate strong data augmentations with shadow counterfeiting to help the network better recognize non-shadow regions and alleviate overfitting. We also devise a simple yet effective label tuning strategy with global-local fusion and shadow-aware filtering to encourage the network to make significant refinements on the noisy labels. We evaluate the performance of SILT by relabeling the test set of the SBU dataset and conducting various experiments. Our results show that even a simple U-Net trained with SILT can outperform all state-of-the-art methods by a large margin. When trained on SBU / UCF / ISTD, our network can successfully reduce the Balanced Error Rate by 25.2% / 36.9% / 21.3% over the best state-of-the-art method.

cs.CV↗

arXiv2023

A Lipschitz Bandits Approach for Continuous Hyperparameter Optimization

One of the most critical problems in machine learning is HyperParameter Optimization (HPO), since choice of hyperparameters has a significant impact on final model performance. Although there are many HPO algorithms, they either have no theoretical guarantees or require strong assumptions. To this end, we introduce BLiE -- a Lipschitz-bandit-based algorithm for HPO that only assumes Lipschitz continuity of the objective function. BLiE exploits the landscape of the objective function to adaptively search over the hyperparameter space. Theoretically, we show that $(i)$ BLiE finds an $ε$-optimal hyperparameter with $\mathcal{O} \left( ε^{-(d_z + β)}\right)$ total budgets, where $d_z$ and $β$ are problem intrinsic; $(ii)$ BLiE is highly parallelizable. Empirically, we demonstrate that BLiE outperforms the state-of-the-art HPO algorithms on benchmark tasks. We also apply BLiE to search for noise schedule of diffusion models. Comparison with the default schedule shows that BLiE schedule greatly improves the sampling speed.

cs.LG↗

arXiv2023

WYWEB: A NLP Evaluation Benchmark For Classical Chinese

To fully evaluate the overall performance of different NLP models in a given domain, many evaluation benchmarks are proposed, such as GLUE, SuperGLUE and CLUE. The fi eld of natural language understanding has traditionally focused on benchmarks for various tasks in languages such as Chinese, English, and multilingua, however, there has been a lack of attention given to the area of classical Chinese, also known as "wen yan wen", which has a rich history spanning thousands of years and holds signifi cant cultural and academic value. For the prosperity of the NLP community, in this paper, we introduce the WYWEB evaluation benchmark, which consists of nine NLP tasks in classical Chinese, implementing sentence classifi cation, sequence labeling, reading comprehension, and machine translation. We evaluate the existing pre-trained language models, which are all struggling with this benchmark. We also introduce a number of supplementary datasets and additional tools to help facilitate further progress on classical Chinese NLU. The github repository is https://github.com/baudzhou/WYWEB.

cs.CL↗

arXiv2023

Multi-legged matter transport: a framework for locomotion on noisy landscapes

While the transport of matter by wheeled vehicles or legged robots can be guaranteed in engineered landscapes like roads or rails, locomotion prediction in complex environments like collapsed buildings or crop fields remains challenging. Inspired by principles of information transmission which allow signals to be reliably transmitted over noisy channels, we develop a ``matter transport" framework demonstrating that non-inertial locomotion can be provably generated over ``noisy" rugose landscapes (heterogeneities on the scale of locomotor dimensions). Experiments confirm that sufficient spatial redundancy in the form of serially-connected legged robots leads to reliable transport on such terrain without requiring sensing and control. Further analogies from communication theory coupled to advances in gaits (coding) and sensor-based feedback control (error detection/correction) can lead to agile locomotion in complex terradynamic regimes.

cs.RO↗

arXiv2023

Convergence Rates of Stochastic Zeroth-order Gradient Descent for Łojasiewicz Functions

We prove convergence rates of Stochastic Zeroth-order Gradient Descent (SZGD) algorithms for Lojasiewicz functions. The SZGD algorithm iterates as \begin{align*} \mathbf{x}_{t+1} = \mathbf{x}_t - η_t \widehat{\nabla} f (\mathbf{x}_t), \qquad t = 0,1,2,3,\cdots , \end{align*} where $f$ is the objective function that satisfies the Łojasiewicz inequality with Łojasiewicz exponent $θ$, $η_t$ is the step size (learning rate), and $ \widehat{\nabla} f (\mathbf{x}_t) $ is the approximate gradient estimated using zeroth-order information only. Our results show that $ \{ f (\mathbf{x}_t) - f (\mathbf{x}_\infty) \}_{t \in \mathbb{N} } $ can converge faster than $ \{ \| \mathbf{x}_t - \mathbf{x}_\infty \| \}_{t \in \mathbb{N} }$, regardless of whether the objective $f$ is smooth or nonsmooth.

math.OC↗

arXiv2023

Topologically assisted optimization for rotor design

We develop and apply a novel shape optimization exemplified for a two-blade rotor with respect to the figure of merit ($FM$). This topologically assisted optimization (TAO) contains two steps. First a global evolutionary optimization is performed for the shape parameters and then a topological analysis reveals the local and global extrema of the objective function directly from the data. This non-dimensional objective function compares the achieved thrust with the required torque. Rotor blades have a decisive contribution to the performance of quadcopters. A two-blade rotor with pre-defined chord length distribution is chosen as the baseline model.The simulation is performed in a moving reference frame with a $k-ω$ turbulence model for the hovering condition.The rotor shape is parameterized by the twist angle distribution.The optimization of this distribution employs a genetic algorithm. The local maxima are distilled from the data using a novel topological analysis inspired by discrete scalar-field topology. We identify one global maximum to be located in the interior of the data and five further local maxima related to errors from non-converged simulations.The interior location of the global optimum suggests that small improvements can be gained from further optimization.The local maxima have a small persistence, i.e., disappear under a small $\varepsilon$ perturbation of the figure of merit values. In other words, the data may be approximated by a smooth mono-modal surrogate model. Thus, the topological data analysis provides valuable insights for optimization and surrogate modeling.

physics.flu-dyn↗

arXiv2023

dame-flame: A Python Library Providing Fast Interpretable Matching for Causal Inference

dame-flame is a Python package for performing matching for observational causal inference on datasets containing discrete covariates. This package implements the Dynamic Almost Matching Exactly (DAME) and Fast Large-Scale Almost Matching Exactly (FLAME) algorithms, which match treatment and control units on subsets of the covariates. The resulting matched groups are interpretable, because the matches are made on covariates, and high-quality, because machine learning is used to determine which covariates are important to match on. DAME solves an optimization problem that matches units on as many covariates as possible, prioritizing matches on important covariates. FLAME approximates the solution found by DAME via a much faster backward feature selection procedure. The package provides several adjustable parameters to adapt the algorithms to specific applications, and can calculate treatment effect estimates after matching. Descriptions of these parameters, details on estimating treatment effects, and further examples, can be found in the documentation at https://almost-matching-exactly.github.io/DAME-FLAME-Python-Package/

cs.LG↗

arXiv2023

Stochastic Zeroth Order Gradient and Hessian Estimators: Variance Reduction and Refined Bias Bounds

We study stochastic zeroth order gradient and Hessian estimators for real-valued functions in $\mathbb{R}^n$. We show that, via taking finite difference along random orthogonal directions, the variance of the stochastic finite difference estimators can be significantly reduced. In particular, we design estimators for smooth functions such that, if one uses $ Θ\left( k \right) $ random directions sampled from the Stiefel's manifold $ \text{St} (n,k) $ and finite-difference granularity $δ$, the variance of the gradient estimator is bounded by $ \mathcal{O} \left( \left( \frac{n}{k} - 1 \right) + \left( \frac{n^2}{k} - n \right) δ^2 + \frac{ n^2 δ^4 }{ k } \right) $, and the variance of the Hessian estimator is bounded by $\mathcal{O} \left( \left( \frac{n^2}{k^2} - 1 \right) + \left( \frac{n^4}{k^2} - n^2 \right) δ^2 + \frac{n^4 δ^4 }{k^2} \right) $. When $k = n$, the variances become negligibly small. In addition, we provide improved bias bounds for the estimators. The bias of both gradient and Hessian estimators for smooth function $f$ is of order $\mathcal{O} \left( δ^2 Γ\right)$, where $δ$ is the finite-difference granularity, and $ Γ$ depends on high order derivatives of $f$. Our results are evidenced by empirical observations.

cs.LG↗

arXiv2023

Optical Transformers

The rapidly increasing size of deep-learning models has caused renewed and growing interest in alternatives to digital computers to dramatically reduce the energy cost of running state-of-the-art neural networks. Optical matrix-vector multipliers are best suited to performing computations with very large operands, which suggests that large Transformer models could be a good target for optical computing. To test this idea, we performed small-scale optical experiments with a prototype accelerator to demonstrate that Transformer operations can run on optical hardware despite noise and errors. Using simulations, validated by our experiments, we then explored the energy efficiency of optical implementations of Transformers and identified scaling laws for model performance with respect to optical energy usage. We found that the optical energy per multiply-accumulate (MAC) scales as $\frac{1}{d}$ where $d$ is the Transformer width, an asymptotic advantage over digital systems. We conclude that with well-engineered, large-scale optical hardware, it may be possible to achieve a $100 \times$ energy-efficiency advantage for running some of the largest current Transformer models, and that if both the models and the optical hardware are scaled to the quadrillion-parameter regime, optical computers could have a $>8,000\times$ energy-efficiency advantage over state-of-the-art digital-electronic processors that achieve 300 fJ/MAC. We analyzed how these results motivate and inform the construction of future optical accelerators along with optics-amenable deep-learning approaches. With assumptions about future improvements to electronics and Transformer quantization techniques (5$\times$ cheaper memory access, double the digital--analog conversion efficiency, and 4-bit precision), we estimated that optical computers' advantage against current 300-fJ/MAC digital processors could grow to $>100,000\times$.

cs.ET↗