5632 research outputs found
Sort by
Through the Lens of A Momentum Microscope: Viewing Light‐Induced Quantum Phenomena in Two‐Dimensional Materials
No full textVan der Waals (vdW) materials at their two-dimensional limit are diverse, flexible, and unique laboratories to study fundamental quantum phenomena and their future applications. Their novel properties rely on their pronounced Coulomb interactions, variety of crystal symmetries and spin-physics, and the ease of incorporation of different vdW materials to form sophisticated heterostructures. In particular, the excited state properties of many two-dimensional semiconductors and semi-metals are relevant for their technological applications, particularly those that can be induced by light. In this paper, we review the recent advances made in studying out-of-equilibrium, light-induced, phenomena in these materials using powerful, surface-sensitive, time-resolved photoemission-based techniques, with a particular emphasis on the emerging multi-dimensional photoemission spectroscopy technique of time-resolved momentum microscopy. We discuss the advances this technique has enabled in studying the nature and dynamics of occupied excited states in these materials. Then, we project for the future research directions opened by these scientific and instrumental advancements, studying the physics of two-dimensional materials and the opportunities to engineer their band-structure and band-topology by laser fields
Cross‐Ray Ultrasound Tomography and Photoacoustic Tomography of Cerebral Hemodynamics in Rodents
No full textRecent advances in functional ultrasound imaging (fUS) and photoacoustic tomography (PAT) offer powerful tools for studying brain function. Complementing each other, fUS and PAT, respectively, measure the cerebral blood flow (CBF) and hemoglobin concentrations, allowing synergistic characterization of cerebral hemodynamics. Here, cross-ray ultrasound tomography (CRUST) and its combination with PAT are presented. CRUST employs a virtual point source from a spherically focused ultrasonic transducer (SFUST) to provide widefield excitation at a 4-kHz pulse repetition frequency. A full-ring-shaped ultrasonic transducer array whose imaging plane is orthogonal to the SFUST's acoustic axis receives scattered ultrasonic waves. Superior to conventional fUS, whose sensitivity to blood flow is angle-dependent and low for perpendicular flow, the crossed transmission and panoramic detection fields of CRUST provide omnidirectional sensitivity to CBF. Using CRUST-PAT, the CBF, oxygen saturation, and hemoglobin concentration changes of the mouse brain during sensory stimulation are measured, with a field of view of ≈7 mm in diameter, spatial resolution of ≈170 µm, and temporal resolution of 200 Hz. The results demonstrate CRUST-PAT as a unique tool for studying cerebral hemodynamics
Sample-Based Bounds for Coherent Risk Measures: Applications to Policy Synthesis and Verification
No full textThe dramatic increase of autonomous systems subject to variable environments has given rise to the pressing need to consider risk in both the synthesis and verification of policies for these systems. This paper aims to address a few problems regarding risk-aware verification and policy synthesis, by first developing a sample-based method to bound the risk measure evaluation of a random variable whose distribution is unknown. These bounds permit us to generate high-confidence verification statements for a large class of robotic systems. Second, we develop a sample-based method to determine solutions to non-convex optimization problems that outperform a large fraction of the decision space of possible solutions. Both sample-based approaches then permit us to rapidly synthesize risk-aware policies that are guaranteed to achieve a minimum level of system performance. To showcase our approach in simulation, we verify a cooperative multi-agent system and develop a risk-aware controller that outperforms the system's baseline controller. We also mention how our approach can be extended to account for any g-entropic risk measure - the subset of coherent risk measures on which we focus
Policy Optimization with Linear Temporal Logic Constraints
No full textWe study the problem of policy optimization (PO) with linear temporal logic (LTL) constraints. The language of LTL allows flexible description of tasks that may be unnatural to encode as a scalar cost function. We consider LTL-constrained PO as a systematic framework, decoupling task specification from policy selection, and an alternative to the standard of cost shaping. With access to a generative model, we develop a model-based approach that enjoys a sample complexity analysis for guaranteeing both task satisfaction and cost optimality (through a reduction to a reachability problem). Empirically, our algorithm can achieve strong performance even in low sample regimes
Explicit Regularization via Regularizer Mirror Descent
No full textDespite perfectly interpolating the training data, deep neural networks (DNNs) can often generalize fairly well, in part due to the "implicit regularization" induced by the learning algorithm. Nonetheless, various forms of regularization, such as "explicit regularization" (via weight decay), are often used to avoid overfitting, especially when the data is corrupted. There are several challenges with explicit regularization, most notably unclear convergence properties. Inspired by convergence properties of stochastic mirror descent (SMD) algorithms, we propose a new method for training DNNs with regularization, called regularizer mirror descent (RMD). In highly overparameterized DNNs, SMD simultaneously interpolates the training data and minimizes a certain potential function of the weights. RMD starts with a standard cost which is the sum of the training loss and a convex regularizer of the weights. Reinterpreting this cost as the potential of an "augmented" overparameterized network and applying SMD yields RMD. As a result, RMD inherits the properties of SMD and provably converges to a point "close" to the minimizer of this cost. RMD is computationally comparable to stochastic gradient descent (SGD) and weight decay, and is parallelizable in the same manner. Our experimental results on training sets with various levels of corruption suggest that the generalization performance of RMD is remarkably robust and significantly better than both SGD and weight decay, which implicitly and explicitly regularize the ℓ₂ norm of the weights. RMD can also be used to regularize the weights to a desired weight vector, which is particularly relevant for continual learning
Finite-Time Regret of Thompson Sampling Algorithms for Exponential Family Multi-Armed Bandits
No full textWe study the regret of Thompson sampling (TS) algorithms for exponential family bandits, where the reward distribution is from a one-dimensional exponential family, which covers many common reward distributions including Bernoulli, Gaussian, Gamma, Exponential, etc. We propose a Thompson sampling algorithm, termed ExpTS, which uses a novel sampling distribution to avoid the under-estimation of the optimal arm. We provide a tight regret analysis for ExpTS, which simultaneously yields both the finite-time regret bound as well as the asymptotic regret bound. In particular, for a K-armed bandit with exponential family rewards, ExpTS over a horizon T is sub-UCB (a strong criterion for the finite-time regret that is problem-dependent), minimax optimal up to a factor √(log K), and asymptotically optimal, for exponential family rewards. Moreover, we propose ExpTS⁺, by adding a greedy exploitation step in addition to the sampling distribution used in ExpTS, to avoid the over-estimation of sub-optimal arms. ExpTS⁺ is an anytime bandit algorithm and achieves the minimax optimality and asymptotic optimality simultaneously for exponential family reward distributions. Our proof techniques are general and conceptually simple and can be easily applied to analyze standard Thompson sampling with specific reward distributions
Quantification of Robotic Surgeries with Vision-Based Deep Learning
No full textSurgery is a high-stakes domain where surgeons must navigate critical anatomical structures and actively avoid potential complications while achieving the main task at hand. Such surgical activity has been shown to affect long-term patient outcomes. To better understand this relationship, whose mechanics remain unknown for the majority of surgical procedures, we hypothesize that the core elements of surgery must first be quantified in a reliable, objective, and scalable manner. We believe this is a prerequisite for the provision of surgical feedback and modulation of surgeon performance in pursuit of improved patient outcomes. To holistically quantify surgeries, we propose a unified deep learning framework, entitled Roboformer, which operates exclusively on videos recorded during surgery to independently achieve multiple tasks: surgical phase recognition (the what of surgery), gesture classification and skills assessment (the how of surgery). We validated our framework on four video-based datasets of two commonly-encountered types of steps (dissection and suturing) within minimally-invasive robotic surgeries. We demonstrated that our framework can generalize well to unseen videos, surgeons, medical centres, and surgical procedures. We also found that our framework, which naturally lends itself to explainable findings, identified relevant information when achieving a particular task. These findings are likely to instill surgeons with more confidence in our framework's behaviour, increasing the likelihood of clinical adoption, and thus paving the way for more targeted surgical feedback
Neural Scene Representation for Locomotion on Structured Terrain
No full textWe propose a learning-based method to reconstruct the local terrain for locomotion with a mobile robot traversing urban environments. Using a stream of depth measurements from the onboard cameras and the robot’s trajectory, the algorithm estimates the topography in the robot’s vicinity. The raw measurements from these cameras are noisy and only provide partial and occluded observations that in many cases do not show the terrain the robot stands on. Therefore, we propose a 3D reconstruction model that faithfully reconstructs the scene, despite the noisy measurements and large amounts of missing data coming from the blind spots of the camera arrangement. The model consists of a 4D fully convolutional network on point clouds that learns the geometric priors to complete the scene from the context and an auto-regressive feedback to leverage spatio-temporal consistency and use evidence from the past. The network can be solely trained with synthetic data, and due to extensive augmentation, it is robust in the real world, as shown in the validation on a quadrupedal robot, ANYmal, traversing challenging settings. We run the pipeline on the robot’s onboard low-power computer using an efficient sparse tensor implementation and show that the proposed method outperforms classical map representations
Generic Lithography Modeling with Dual-band Optics-Inspired Neural Networks
No full textLithography simulation is a critical step in VLSI design and optimization for manufacturability. Existing solutions for highly accurate lithography simulation with rigorous models are computationally expensive and slow, even when equipped with various approximation techniques. Recently, machine learning has provided alternative solutions for lithography simulation tasks such as coarse-grained edge placement error regression and complete contour prediction. However, the impact of these learning-based methods has been limited due to restrictive usage scenarios or low simulation accuracy. To tackle these concerns, we introduce an dual-band optics-inspired neural network design that considers the optical physics underlying lithography. To the best of our knowledge, our approach yields the first published via/metal layer contour simulation at 1nm^2/pixel resolution with any tile size. Compared to previous machine learning based solutions, we demonstrate that our framework can be trained much faster and offers a significant improvement on efficiency and image quality with 20X smaller model size. We also achieve 85X simulation speedup over traditional lithography simulator with 1% accuracy loss
Sequential Prefrontal and Temporoparietal Repetitive Transcranial Magnetic Stimulation (rTMS) for Treatment of Tinnitus With and Without Comorbid Depression: A Case Series and Systematic Review
No full textBackground. Tinnitus distress is related to both the loudness and intrusiveness of the tinnitus percept. Treatment approaches targeting both attentional/limbic and auditory systems may better alleviate tinnitus distress than approaches targeting the auditory system alone.
Materials and Methods. Ten subjects with chronic tinnitus received sequential rTMS treatment involving: 1) excitatory stimulation administered to the left dorsolateral prefrontal cortex (DLPFC) or inhibitory stimulation administered to the right DLPFC, followed by 2) inhibitory stimulation administered to primary auditory cortex (Heschel's gyrus or HG). A systematic literature review was performed to evaluate the existing literature on sequential repetitive Transcranial Magnetic Stimulation (rTMS) treatment approaches for tinnitus. Results of the case series are interpreted in the context of tinnitus neurobiology and the extant literature.
Results. Subjects experienced a significant decrease (average 21.7%) in symptoms on the Tinnitus Functional Index (TFI). Those with tinnitus alone experienced a greater mean symptom reduction than those with comorbid MDD (27.7 vs. 17.0%, respectively). Adverse effects were transient and minor. Literature review confirmed that sequential approaches had some advantages compared to single site rTMS; in general, the addition of 1 Hz treatment at DLPFC was superior to single site rTMS in the short term (1–12 weeks), while the addition of 20 Hz treatment at DLPFC appeared superior in the long term (90–180 days).
Conclusions. Sequential rTMS approaches for the treatment of tinnitus—particularly those administering low-frequency treatment at left DLPFC—merit further investigation