BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Tianyu Wang

Tianyu Wang contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Machine LearningComputer VisionRoboticsEmerging Technologiesphysics.opticsArtificial IntelligenceComputation and LanguageNeural and Evolutionary ComputingBiological Physicscond-mat.dis-nncond-mat.softcs.CYDatabasesGraphicsmath.NAmath.OCnucl-thNumerical Analysis

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
29works
0followers
18topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

29 published item(s)

preprint2026arXiv

Can LLMs Think Like Consumers? Benchmarking Crowd-Level Reaction Reconstruction with ConsumerSimBench

LLMs are increasingly used as ``digital consumers'' to simulate public opinion, pre-test marketing decisions, and anticipate audience response. However, existing evaluations rarely ask whether a model can reconstruct the concrete reaction patterns that real consumers surface in public discourse. We introduce ConsumerSimBench, a benchmark built from 1,553 real Chinese social-media topics and 23,122 atomic, rule-audited criteria spanning four reaction families. Rather than scoring open-ended generations with a holistic preference judge, ConsumerSimBench decomposes each task into auditable yes-no decisions over concrete reaction points, raising three-judge agreement from 65.8% to 92.1% with 98.4% agreement between pointwise judge decisions and human-majority labels. Across 13 frontier generators, the strongest model, Gemini-3.1-Pro, covers only 47.8% of real reaction criteria, while GPT-5.2 and Claude-4.6 trail far behind despite their strength on technical benchmarks. The failures reveal a sharp gap between technical-benchmark performance and socially grounded consumer intuition. A direct structured reasoning prompt decreases coverage, while a generate--reflect multi-agent pipeline improves MiMo-V2.5-Pro from 32.9% to 37.6% on a subset. ConsumerSimBench reframes consumer simulation as a forecasting problem over real public-discourse reactions, showing that frontier LLMs remain far from reliably predicting what consumers will actually care about in high-context Chinese consumer discourse.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

DecisionLLM: Large Language Models for Long Sequence Decision Exploration

Long-sequence decision-making, which is usually addressed through reinforcement learning (RL), is a critical component for optimizing strategic operations in dynamic environments, such as real-time bidding in computational advertising. The Decision Transformer (DT) introduced a powerful paradigm by framing RL as an autoregressive sequence modeling problem. Concurrently, Large Language Models (LLMs) have demonstrated remarkable success in complex reasoning and planning tasks. This inspires us whether LLMs, which share the same Transformer foundation, but operate at a much larger scale, can unlock new levels of performance in long-horizon sequential decision-making problem. This work investigates the application of LLMs to offline decision making tasks. A fundamental challenge in this domain is the LLMs' inherent inability to interpret continuous values, as they lack a native understanding of numerical magnitude and order when values are represented as text strings. To address this, we propose treating trajectories as a distinct modality. By learning to align trajectory data with natural language task descriptions, our model can autoregressively predict future decisions within a cohesive framework we term DecisionLLM. We establish a set of scaling laws governing this paradigm, demonstrating that performance hinges on three factors: model scale, data volume, and data quality. In offline experimental benchmarks and bidding scenarios, DecisionLLM achieves strong performance. Specifically, DecisionLLM-3B outperforms the traditional Decision Transformer (DT) by 69.4 on Maze2D umaze-v1 and by 0.085 on AuctionNet. It extends the AIGB paradigm and points to promising directions for future exploration in online bidding.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

In-medium nucleon-nucleon cross sections from relativistic ab initio calculations

The in-medium nucleon-nucleon scattering cross section is a pivotal quantity for studying the medium effects of strong interaction, and its precise knowledge is critical for understanding the equation of state for dense matter, intermediate-energy heavy-ion collision dynamics, and related phenomena. In this work, we perform a microscopic investigation of in-medium nucleon-nucleon scattering cross sections, by utilizing the relativistic Brueckner-Hartree-Fock (RBHF) theory with the Bonn potential. The fully incorporation of both positive- and negative-energy states in the RBHF solutions allows us to determine the single-particle potentials, the effective G matrix, and the scattering cross section uniquely. The momentum, density, and isospin dependence of the cross section for pp, nn, and np scattering are studied in detail. Our results provide a solid foundation for future parametrization studies of multiparameter dependency of total scattering cross sections.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM

Standard differential privacy imposes uniform privacy constraints across all features, overlooking the inherent distinction between sensitive and insensitive features in practice. In this paper, we introduce a relaxed definition of differential privacy that accounts for such heterogeneity, allowing certain features to be treated as insensitive even when correlated with sensitive ones. We propose a correlation-aware framework, $\textsf{CorrDP}$, which relaxes privacy for insensitive features while accounting for their correlations with sensitive features, with the correlations quantified using total variation distance. We design algorithms for differentially private empirical risk minimization (DP-ERM) under the $\textsf{CorrDP}$ framework, incorporating distance-dependent noise into gradients for improved theoretical utility guarantees. When the correlation distance is unknown, we estimate it from the dataset and show that it achieves a comparable privacy-utility guarantee. We perform experiments on synthetic and real-world datasets and show that $\textsf{CorrDP}$-based DP-ERM algorithms consistently outperform the standard DP framework in the presence of insensitive features.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Measurement-Adapted Eigentask Representations for Photon-Limited Optical Readout

Optical readout in low-light imaging is fundamentally limited by measurement noise, including photon shot noise, detector noise, and quantization error. In this regime, downstream inference depends not only on the optical front end, but also on how noisy high-dimensional sensor measurements are represented before classification or decision-making. Here we show that eigentasks provide a measurement-adapted representation for optical sensor outputs by ordering readout features according to their resolvability under noise. Using experimental data from a lens-based optical imaging system and a reanalysis of published data from a single-photon-detection neural network, we find that eigentask representations frequently outperform standard baselines including principal component analysis and filtering-based compression. The advantage is most pronounced in photon-limited, few-shot, and higher-difficulty classification regimes. In few-shot MPEG-7 classification, for example, the advantage over other methods reaches about 10 percentage points as the number of classes increases. In these settings, eigentasks yield more informative low-dimensional features and improve sample-efficient downstream learning. These results identify measurement-adapted representation as a promising strategy for optical inference when photon budget, acquisition time, and task complexity are constrained.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Physical Foundation Models: Fixed hardware implementations of large-scale neural networks

Foundation models are deep neural networks (such as GPT-5, Gemini~3, and Opus~4) trained on large datasets that can perform diverse downstream tasks -- text and code generation, question answering, summarization, image classification, and so on. The philosophy of foundation models is to put effort into a single, large (${\sim}10^{12}$-parameter) general-purpose model that can be adapted to many downstream tasks with no or minimal additional training. We argue that the rise of foundation models presents an opportunity for hardware engineers: in contrast to when different models were used for different tasks, it now makes sense to build special-purpose, fixed hardware implementations of neural networks, manufactured and released at the roughly 1-year cadence of major new foundation-model versions. Beyond conventional digital-electronic inference hardware with read-only weight memory, we advocate a more radical re-thinking: hardware in which the neural network is realized directly at the level of the physical design and operates via the hardware's natural physical dynamics -- \textit{Physical Foundation Models} (PFMs). PFMs could enable orders-of-magnitude advantages in energy efficiency, speed, and parameter density. For ${\sim}10^{12}$-parameter models, this would both reduce the high energy burden of AI in datacenters and enable AI in edge devices that today are power-constrained to far smaller models. PFMs could also enable inference hardware for models much larger than current ones: $10^{15}$- or even $10^{18}$-parameter PFMs seem plausible by some measures. We present back-of-the-envelope calculations illustrating PFM scaling using an optical example -- a 3D nanostructured glass medium -- and discuss prospects in nanoelectronics and other physical platforms. We conclude with the major research challenges that must be resolved for trillion-parameter PFMs and beyond to become reality.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SAGE: Hierarchical LLM-Based Literary Evaluation through Ontology-Grounded Interpretive Dimensions

Evaluating literary quality requires assessing interpretive dimensions such as cultural representation, emotional depth, and philosophical sophistication that resist straightforward computational measurement. We introduce SAGE, a hierarchical evaluation framework that decomposes literary quality into ontology-grounded interpretive dimensions assessed through structured large language model evaluation with multi-round iterative reflection and independent validation. We validate the framework on 100 short stories (50 canonical works, 30 pulp fiction, 20 LLM-generated narratives) across three analytical layers (cultural, emotional-psychological, existential-philosophical) using dual-mode assessment. Across 600 evaluations, the framework achieves 98.8% score convergence and greater than 94% inter-rater agreement, with near-perfect mode invariance between content-based and metadata-based evaluation. Statistical analysis reveals a consistent genre hierarchy (Canonical > Pulp > LLM, all p<0.001) with layer-specific discrimination: cultural critique and philosophical depth exhibit very large effect sizes (Cohen's d>2.4), while emotional representation shows smaller gaps (d=1.68), suggesting that affective patterns are more learnable from training data than critical stance or philosophical depth. Cross-layer correlations (r=0.649-0.683) confirm the three dimensions capture empirically distinguishable quality facets. These findings demonstrate that theory-driven LLM evaluation can achieve measurement-grade reliability and support systematic identification of where current generative models fall short of human literary production, with direct implications for scalable automated evaluation of open-ended text generation.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

OmniVCus: Feedforward Subject-driven Video Customization with Multimodal Control Conditions

Existing feedforward subject-driven video customization methods mainly study single-subject scenarios due to the difficulty of constructing multi-subject training data pairs. Another challenging problem that how to use the signals such as depth, mask, camera, and text prompts to control and edit the subject in the customized video is still less explored. In this paper, we first propose a data construction pipeline, VideoCus-Factory, to produce training data pairs for multi-subject customization from raw videos without labels and control signals such as depth-to-video and mask-to-video pairs. Based on our constructed data, we develop an Image-Video Transfer Mixed (IVTM) training with image editing data to enable instructive editing for the subject in the customized video. Then we propose a diffusion Transformer framework, OmniVCus, with two embedding mechanisms, Lottery Embedding (LE) and Temporally Aligned Embedding (TAE). LE enables inference with more subjects by using the training subjects to activate more frame embeddings. TAE encourages the generation process to extract guidance from temporally aligned control signals by assigning the same frame embeddings to the control and noise tokens. Experiments demonstrate that our method significantly surpasses state-of-the-art methods in both quantitative and qualitative evaluations. Video demos are at our project page: https://caiyuanhao1998.github.io/project/OmniVCus/. Our code, models, data are released at https://github.com/caiyuanhao1998/Open-OmniVCus

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A general locomotion control framework for multi-legged locomotors

Serially connected robots are promising candidates for performing tasks in confined spaces such as search-and-rescue in large-scale disasters. Such robots are typically limbless, and we hypothesize that the addition of limbs could improve mobility. However, a challenge in designing and controlling such devices lies in the coordination of high-dimensional redundant modules in a way that improves mobility. Here we develop a general framework to control serially connected multi-legged robots. Specifically, we combine two approaches to build a general shape control scheme which can provide baseline patterns of self-deformation (&#34;gaits&#34;) for effective locomotion in diverse robot morphologies. First, we take inspiration from a dimensionality reduction and a biological gait classification scheme to generate cyclic patterns of body deformation and foot lifting/lowering, which facilitate generation of arbitrary substrate contact patterns. Second, we use geometric mechanics methods to facilitates identification of optimal phasing of these undulations to maximize speed and/or stability. Our scheme allows the development of effective gaits in multi-legged robots locomoting on flat frictional terrain with diverse number of limbs (4, 6, 16, and even 0 limbs) and body actuation capabilities (including sidewinding gaits on limbless devices). By properly coordinating the body undulation and the leg placement, our framework combines the advantages of both limbless robots (modularity) and legged robots (mobility). We expect that our framework can provide general control schemes for the rapid deployment of general multi-legged robots, paving the ways toward machines that can traverse complex environments under real-life conditions.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Coordinating tiny limbs and long bodies: geometric mechanics of diverse undulatory lizard locomotion

Although typically possessing four limbs and short bodies, lizards have evolved a diversity of body plans, from short-bodied and fully-limbed to elongate and nearly limbless. Such diversity in body morphology is hypothesized as adaptations to locomotion cluttered terrestrial environments, but the mode of propulsion -- e.g., the use of body and/or limbs to interact with the substrate -- and potential body/limb coordination remain unstudied. Here, we use biological experiments, a geometric theory of locomotion, and robophysical experiments to comparatively and systematically investigate such dynamics in a diverse sample of lizard morphologies. Locomotor field studies in short-limb, elongated lizards (Brachymeles) and laboratory studies of full-limbed lizards (Uma scoparia and Sceloporus olivaceus) and a limbless laterally undulating organism (Chionactis occipitalis) reveal that the body wave dynamics can be described by a combination of traveling and standing waves; the ratio of the amplitudes of these components is inversely related to limb length. We use geometric theory to analyze and explain the wave dynamics and body-leg coordination observations; the theory predicts that leg thrust modulates the body weight distribution and self-propulsion generation mechanism, which in turn facilitates the choice of body waves. We test our hypothesis in biological experiments by inducing the use of traveling wave in stereotyped lizards by modulating the ground penetration resistance, as well as in controlled non-biological experiments involving an undulating limbed robophysical model. Our models could be valuable in understanding functional constraints on the evolutionary process of elongation and limb reduction in lizards, as well as advancing robot designs.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

DMF-Net: Dual-Branch Multi-Scale Feature Fusion Network for copy forgery identification of anti-counterfeiting QR code

Anti-counterfeiting QR codes are widely used in people&#39;s work and life, especially in product packaging. However, the anti-counterfeiting QR code has the risk of being copied and forged in the circulation process. In reality, copying is usually based on genuine anti-counterfeiting QR codes, but the brands and models of copiers are diverse, and it is extremely difficult to determine which individual copier the forged anti-counterfeiting code come from. In response to the above problems, this paper proposes a method for copy forgery identification of anti-counterfeiting QR code based on deep learning. We first analyze the production principle of anti-counterfeiting QR code, and convert the identification of copy forgery to device category forensics, and then a Dual-Branch Multi-Scale Feature Fusion network is proposed. During the design of the network, we conducted a detailed analysis of the data preprocessing layer, single-branch design, etc., combined with experiments, the specific structure of the dual-branch multi-scale feature fusion network is determined. The experimental results show that the proposed method has achieved a high accuracy of copy forgery identification, which exceeds the current series of methods in the field of image forensics.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

From the Greene--Wu Convolution to Gradient Estimation over Riemannian Manifolds

Over a complete Riemannian manifold of finite dimension, Greene and Wu introduced a convolution, known as Greene-Wu (GW) convolution. In this paper, we study properties of the GW convolution and apply it to non-Euclidean machine learning problems. In particular, we derive a new formula for how the curvature of the space would affect the curvature of the function through the GW convolution. Also, following the study of the GW convolution, a new method for gradient estimation over Riemannian manifolds is introduced.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Instance Shadow Detection with A Single-Stage Detector

This paper formulates a new problem, instance shadow detection, which aims to detect shadow instance and the associated object instance that cast each shadow in the input image. To approach this task, we first compile a new dataset with the masks for shadow instances, object instances, and shadow-object associations. We then design an evaluation metric for quantitative evaluation of the performance of instance shadow detection. Further, we design a single-stage detector to perform instance shadow detection in an end-to-end manner, where the bidirectional relation learning module and the deformable maskIoU head are proposed in the detector to directly learn the relation between shadow instances and object instances and to improve the accuracy of the predicted masks. Finally, we quantitatively and qualitatively evaluate our method on the benchmark dataset of instance shadow detection and show the applicability of our method on light direction estimation and photo editing.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Latent Policies for Adversarial Imitation Learning

This paper considers learning robot locomotion and manipulation tasks from expert demonstrations. Generative adversarial imitation learning (GAIL) trains a discriminator that distinguishes expert from agent transitions, and in turn use a reward defined by the discriminator output to optimize a policy generator for the agent. This generative adversarial training approach is very powerful but depends on a delicate balance between the discriminator and the generator training. In high-dimensional problems, the discriminator training may easily overfit or exploit associations with task-irrelevant features for transition classification. A key insight of this work is that performing imitation learning in a suitable latent task space makes the training process stable, even in challenging high-dimensional problems. We use an action encoder-decoder model to obtain a low-dimensional latent action space and train a LAtent Policy using Adversarial imitation Learning (LAPAL). The encoder-decoder model can be trained offline from state-action pairs to obtain a task-agnostic latent action representation or online, simultaneously with the discriminator and generator training, to obtain a task-aware latent action representation. We demonstrate that LAPAL training is stable, with near-monotonic performance improvement, and achieves expert performance in most locomotion and manipulation tasks, while a GAIL baseline converges slower and does not achieve expert performance in high-dimensional environments.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Locomotion without force, and impulse via dissipation: Robotic swimming in curved space via geometric phase

Locomotion by shape changes (spermatozoon swimming, snake slithering, bird flapping) or gas expulsion (rocket firing) is assumed to require environmental interaction, due to conservation of momentum. As first noted in (Wisdom, 2003) and later in (Guéron, 2009) and (Avron et al, 2006), in curved space or spacetime the non-commutativity of translations permits translation without momentum exchange, just as falling cats and lizards can self-deform to reorient in flat space without environmental interaction. Translation in curved space can occur not only in gravitationally induced curved spacetime (where translation is predicted to be on the order of $10^{-23}$ m per gait cycle) but also in the curved surfaces encountered by locomotors in real-world environments. Here we show that a precision robophysical apparatus consisting of motors driven on curved tracks (and thereby confined to a spherical surface without a solid substrate) can self-propel without environmental momentum exchange (impulse) via shape changes that can generate gauge potentials that manifest as translations. Our system produces shape changes comparable to the environment&#39;s inverse curvatures and generates from zero momentum forward movement of $10^{-1}$ cm per gait cycle even while resisted by weak gravitational and frictional forces. Dissipation via friction eventually arrests the robot but also imbues it with momentum which can be released upon a cessation of shape changes. This work demonstrates how the interaction between environmental curvature, active driving and geometric phases yields rich, exotic phenomena.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Spatially Invariant Unsupervised 3D Object-Centric Learning and Scene Decomposition

We tackle the problem of object-centric learning on point clouds, which is crucial for high-level relational reasoning and scalable machine intelligence. In particular, we introduce a framework, SPAIR3D, to factorize a 3D point cloud into a spatial mixture model where each component corresponds to one object. To model the spatial mixture model on point clouds, we derive the Chamfer Mixture Loss, which fits naturally into our variational training pipeline. Moreover, we adopt an object-specification scheme that describes each object&#39;s location relative to its local voxel grid cell. Such a scheme allows SPAIR3D to model scenes with an arbitrary number of objects. We evaluate our method on the task of unsupervised scene decomposition. Experimental results demonstrate that SPAIR3D has strong scalability and is capable of detecting and segmenting an unknown number of objects from a point cloud in an unsupervised manner.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Towards Fundamental Limits of Multi-armed Bandits with Random Walk Feedback

In this paper, we consider a new Multi-Armed Bandit (MAB) problem where arms are nodes in an unknown and possibly changing graph, and the agent (i) initiates random walks over the graph by pulling arms, (ii) observes the random walk trajectories, and (iii) receives rewards equal to the lengths of the walks. We provide a comprehensive understanding of this problem by studying both the stochastic and the adversarial setting. We show that this problem is not easier than a standard MAB in an information theoretical sense, although additional information is available through random walk trajectories. Behaviors of bandit algorithms on this problem are also studied.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

An optical neural network using less than 1 photon per multiplication

Deep learning has rapidly become a widespread tool in both scientific and commercial endeavors. Milestones of deep learning exceeding human performance have been achieved for a growing number of tasks over the past several years, across areas as diverse as game-playing, natural-language translation, and medical-image analysis. However, continued progress is increasingly hampered by the high energy costs associated with training and running deep neural networks on electronic processors. Optical neural networks have attracted attention as an alternative physical platform for deep learning, as it has been theoretically predicted that they can fundamentally achieve higher energy efficiency than neural networks deployed on conventional digital computers. Here, we experimentally demonstrate an optical neural network achieving 99% accuracy on handwritten-digit classification using ~3.2 detected photons per weight multiplication and ~90% accuracy using ~0.64 photons (~$2.4 \times 10^{-19}$ J of optical energy) per weight multiplication. This performance was achieved using a custom free-space optical processor that executes matrix-vector multiplications in a massively parallel fashion, with up to ~0.5 million scalar (weight) multiplications performed at the same time. Using commercially available optical components and standard neural-network training methods, we demonstrated that optical neural networks can operate near the standard quantum limit with extremely low optical powers and still achieve high accuracy. Our results provide a proof-of-principle for low-optical-power operation, and with careful system design including the surrounding electronics used for data storage and control, open up a path to realizing optical processors that require only $10^{-16}$ J total energy per scalar multiplication -- which is orders of magnitude more efficient than current digital processors.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems

Deep neural networks have become a pervasive tool in science and engineering. However, modern deep neural networks&#39; growing energy requirements now increasingly limit their scaling and broader use. We propose a radical alternative for implementing deep neural network models: Physical Neural Networks. We introduce a hybrid physical-digital algorithm called Physics-Aware Training to efficiently train sequences of controllable physical systems to act as deep neural networks. This method automatically trains the functionality of any sequence of real physical systems, directly, using backpropagation, the same technique used for modern deep neural networks. To illustrate their generality, we demonstrate physical neural networks with three diverse physical systems-optical, mechanical, and electrical. Physical neural networks may facilitate unconventional machine learning hardware that is orders of magnitude faster and more energy efficient than conventional electronic processors.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Episodic Linear Quadratic Regulators with Low-rank Transitions

Linear Quadratic Regulators (LQR) achieve enormous successful real-world applications. Very recently, people have been focusing on efficient learning algorithms for LQRs when their dynamics are unknown. Existing results effectively learn to control the unknown system using number of episodes depending polynomially on the system parameters, including the ambient dimension of the states. These traditional approaches, however, become inefficient in common scenarios, e.g., when the states are high-resolution images. In this paper, we propose an algorithm that utilizes the intrinsic system low-rank structure for efficient learning. For problems of rank-$m$, our algorithm achieves a $K$-episode regret bound of order $\widetilde{O}(m^{3/2} K^{1/2})$. Consequently, the sample complexity of our algorithm only depends on the rank, $m$, rather than the ambient dimension, $d$, which can be orders-of-magnitude larger.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference

A classical problem in causal inference is that of matching, where treatment units need to be matched to control units based on covariate information. In this work, we propose a method that computes high quality almost-exact matches for high-dimensional categorical datasets. This method, called FLAME (Fast Large-scale Almost Matching Exactly), learns a distance metric for matching using a hold-out training data set. In order to perform matching efficiently for large datasets, FLAME leverages techniques that are natural for query processing in the area of database management, and two implementations of FLAME are provided: the first uses SQL queries and the second uses bit-vector techniques. The algorithm starts by constructing matches of the highest quality (exact matches on all covariates), and successively eliminates variables in order to match exactly on as many variables as possible, while still maintaining interpretable high-quality matches and balance between treatment and control groups. We leverage these high quality matches to estimate conditional average treatment effects (CATEs). Our experiments show that FLAME scales to huge datasets with millions of observations where existing state-of-the-art methods fail, and that it achieves significantly better performance than other matching methods.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Inverse reinforcement learning for autonomous navigation via differentiable semantic mapping and planning

This paper focuses on inverse reinforcement learning for autonomous navigation using distance and semantic category observations. The objective is to infer a cost function that explains demonstrated behavior while relying only on the expert&#39;s observations and state-control trajectory. We develop a map encoder, that infers semantic category probabilities from the observation sequence, and a cost encoder, defined as a deep neural network over the semantic features. Since the expert cost is not directly observable, the model parameters can only be optimized by differentiating the error between demonstrated controls and a control policy computed from the cost estimate. We propose a new model of expert behavior that enables error minimization using a closed-form subgradient computed only over a subset of promising states via a motion planning algorithm. Our approach allows generalizing the learned behavior to new environments with new spatial configurations of the semantic categories. We analyze the different components of our model in a minigrid environment. We also demonstrate that our approach learns to follow traffic rules in the autonomous driving CARLA simulator by relying on semantic observations of buildings, sidewalks, and road lanes.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Reconstruction of Backbone Curves for Snake Robots

Snake robots composed of alternating single-axis pitch and yaw joints have many internal degrees of freedom, which make them capable of versatile three-dimensional locomotion. In motion planning process, snake robot motions are often designed kinematically by a chronological sequence of continuous backbone curves that capture desired macroscopic shapes of the robot. However, as the geometric arrangement of single-axis rotary joints creates constraints on the rotations in the robot, it is challenging for the robot to reconstruct an arbitrary 3D curve. When the robot configuration does not accurately achieve the desired shapes defined by these backbone curves, the robot can have unexpected contacts with the environment, such that the robot does not achieve the desired motion. In this work, we propose a method for snake robots to reconstruct desired backbone curves by posing an optimization problem that exploits the robot&#39;s geometric structure. We verified that our method enables fast and accurate curve-configuration conversions through its applications to commonly used 3D gaits. We also demonstrated via robot experiments that 1) our method results in smooth locomotion on the robot; 2) our method allows the robot to approach the numerically predicted locomotive performance of a sequence of continuous backbone curve.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Towards Practical Lipschitz Bandits

Stochastic Lipschitz bandit algorithms balance exploration and exploitation, and have been used for a variety of important task domains. In this paper, we present a framework for Lipschitz bandit methods that adaptively learns partitions of context- and arm-space. Due to this flexibility, the algorithm is able to efficiently optimize rewards and minimize regret, by focusing on the portions of the space that are most relevant. In our analysis, we link tree-based methods to Gaussian processes. In light of our analysis, we design a novel hierarchical Bayesian model for Lipschitz bandit problems. Our experiments show that our algorithms can achieve state-of-the-art performance in challenging real-world tasks such as neural network hyperparameter tuning.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Bandits for BMO Functions

We study the bandit problem where the underlying expected reward is a Bounded Mean Oscillation (BMO) function. BMO functions are allowed to be discontinuous and unbounded, and are useful in modeling signals with infinities in the do-main. We develop a toolset for BMO bandits, and provide an algorithm that can achieve poly-log $δ$-regret -- a regret measured against an arm that is optimal after removing a $δ$-sized portion of the arm space.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Instance Shadow Detection

Instance shadow detection is a brand new problem, aiming to find shadow instances paired with object instances. To approach it, we first prepare a new dataset called SOBA, named after Shadow-OBject Association, with 3,623 pairs of shadow and object instances in 1,000 photos, each with individual labeled masks. Second, we design LISA, named after Light-guided Instance Shadow-object Association, an end-to-end framework to automatically predict the shadow and object instances, together with the shadow-object associations and light direction. Then, we pair up the predicted shadow and object instances, and match them with the predicted shadow-object associations to generate the final results. In our evaluations, we formulate a new metric named the shadow-object average precision to measure the performance of our results. Further, we conducted various experiments and demonstrate our method&#39;s applicability on light direction estimation and photo editing.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Learning Navigation Costs from Demonstration in Partially Observable Environments

This paper focuses on inverse reinforcement learning (IRL) to enable safe and efficient autonomous navigation in unknown partially observable environments. The objective is to infer a cost function that explains expert-demonstrated navigation behavior while relying only on the observations and state-control trajectory used by the expert. We develop a cost function representation composed of two parts: a probabilistic occupancy encoder, with recurrent dependence on the observation sequence, and a cost encoder, defined over the occupancy features. The representation parameters are optimized by differentiating the error between demonstrated controls and a control policy computed from the cost encoder. Such differentiation is typically computed by dynamic programming through the value function over the whole state space. We observe that this is inefficient in large partially observable environments because most states are unexplored. Instead, we rely on a closed-form subgradient of the cost-to-go obtained only over a subset of promising states via an efficient motion-planning algorithm such as A* or RRT. Our experiments show that our model exceeds the accuracy of baseline IRL algorithms in robot navigation tasks, while substantially improving the efficiency of training and test-time inference.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Learning Navigation Costs from Demonstration with Semantic Observations

This paper focuses on inverse reinforcement learning (IRL) for autonomous robot navigation using semantic observations. The objective is to infer a cost function that explains demonstrated behavior while relying only on the expert&#39;s observations and state-control trajectory. We develop a map encoder, which infers semantic class probabilities from the observation sequence, and a cost encoder, defined as deep neural network over the semantic features. Since the expert cost is not directly observable, the representation parameters can only be optimized by differentiating the error between demonstrated controls and a control policy computed from the cost estimate. The error is optimized using a closed-form subgradient computed only over a subset of promising states via a motion planning algorithm. We show that our approach learns to follow traffic rules in the autonomous driving CARLA simulator by relying on semantic observations of cars, sidewalks and road lanes.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

SAC-Net: Spatial Attenuation Context for Salient Object Detection

This paper presents a new deep neural network design for salient object detection by maximizing the integration of local and global image context within, around, and beyond the salient objects. Our key idea is to adaptively propagate and aggregate the image context features with variable attenuation over the entire feature maps. To achieve this, we design the spatial attenuation context (SAC) module to recurrently translate and aggregate the context features independently with different attenuation factors and then to attentively learn the weights to adaptively integrate the aggregated context features. By further embedding the module to process individual layers in a deep network, namely SAC-Net, we can train the network end-to-end and optimize the context features for detecting salient objects. Compared with 29 state-of-the-art methods, experimental results show that our method performs favorably over all the others on six common benchmark data, both quantitatively and visually.

0 citations0 reviewsNo code linkedOpen access