arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Ben Graham

10 papers · 2011–2023

2011: 12012: 02013: 02014: 02015: 22016: 02017: 02018: 02019: 12020: 02021: 12022: 22023: 3
2011peak 3 in 20232023

Frequent co-authors

Papers

showing 1–10 of 10 RSS

LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference

Ben Graham, Alaaeldin El-Nouby, Hugo Touvron +4 more

cs.CV · · cited 1,079× · arXiv ↗ · more like this ·

We design a family of image classification architectures that optimize the trade-off between accuracy and efficiency in a high-speed regime. Our work exploits recent findings in attention-based architectures, which are competitive on highly parallel processing hardware. We revisit principles from the extensive literature on convolutional neural networks to apply them to transformers, in particular activation maps with decreasing resolutions. We also introduce the attention bias, a new way to integrate positional information in vision transformers. As a result, we propose LeVIT: a hybrid neural network for fast inference image classification. We consider different measures of efficiency on different hardware platforms, so as to best reflect a wide range of application scenarios. Our extensive experiments empirically validate our technical choices and show they are suitable to most…

Sparse 3D convolutional neural networks

Ben Graham

cs.CV · · cited 226× · arXiv ↗ · more like this ·

We have implemented a convolutional neural network designed for processing sparse three-dimensional input data. The world we live in is three dimensional so there are a large number of potential applications including 3D object recognition and analysis of space-time objects. In the quest for efficiency, we experiment with CNNs on the 2D triangular-lattice and 3D tetrahedral-lattice.

PerspectiveNet: A Scene-consistent Image Generator for New View Synthesis in Real Indoor Environments

David Novotný, Ben Graham, Jeremy Reizenstein

Computer Vision and Pattern Recognition · Neural Information Processing Systems · · cited 20× · OpenAlex ↗ · more like this ·

Given a set of a reference RGBD views of an indoor environment, and a new viewpoint, our goal is to predict the view from that location. Prior work on new-view generation has predominantly focused on significantly constrained scenarios, typically involving artificially rendered views of isolated CAD models. Here we tackle a much more challenging version of the problem. We devise an approach that exploits known geometric properties of the scene (per-frame camera extrinsics and depth) in order to warp reference views into the new ones. The defects in the generated views are handled by a novel RGBD inpainting network, PerspectiveNet, that is fine-tuned for a given scene in order to obtain images that are geometrically consistent with all the views in the scene camera system. Experiments conducted on the ScanNet and SceneNet datasets reveal performance superior to strong baselines.

Analysis of the Effect of Human Presence on a Wireless Sensor Network

Ben Graham, Christos Tachtatzis, Fabio Di Franco +4 more

Computer Networks and Communications · International Journal of Ambient Computing and Intelligence · · cited 11× · DOI ↗ · more like this ·

Wireless Sensor Networks (WSNs) are gaining an increasing industry wide adoption. However, there remain major challenges such as network dimensioning and node placement especially in Built Environment Networks (BENs). Decisions on the node placement, orientation, and the number of nodes to cover the area of interest are usually ad-hoc. Ray tracing tools are traditionally employed to predict RF signal propagation; however, such tools are primarily intended for outdoor environments. RF signal propagation varies greatly indoors due to building materials and infrastructure, obstacles, node placement, antenna orientation and human presence. Because of the complexity of signal prediction, these factors are usually ignored or given little weight when such networks are analyzed. The paper’s results show the effects of the building size and layout, building materials, human presence and mobility…

Efficient batchwise dropout training using submatrices

Ben Graham, Jeremy Reizenstein, Leigh Robinson

cs.NE · · cited 7× · arXiv ↗ · more like this ·

Dropout is a popular technique for regularizing artificial neural networks. Dropout networks are generally trained by minibatch gradient descent with a dropout mask turning off some of the units---a different pattern of dropout is applied to every sample in the minibatch. We explore a very simple alternative to the dropout mask. Instead of masking dropped out units by setting them to zero, we perform matrix multiplication using a submatrix of the weight matrix---unneeded hidden units are never calculated. Performing dropout batchwise, so that one pattern of dropout is used for each sample in a minibatch, we can substantially reduce training times. Batchwise dropout can be used with fully-connected and convolutional neural networks.

Replay: Multi-modal Multi-view Acted Videos for Casual Holography

Roman Shapovalov, Yanir Kleiman, Ignacio Rocco +6 more

cs.CV · · cited 5× · arXiv ↗ · more like this ·

We introduce Replay, a collection of multi-view, multi-modal videos of humans interacting socially. Each scene is filmed in high production quality, from different viewpoints with several static cameras, as well as wearable action cameras, and recorded with a large array of microphones at different positions in the room. Overall, the dataset contains over 4000 minutes of footage and over 7 million timestamped high-resolution frames annotated with camera poses and partially with foreground masks. The Replay dataset has many potential applications, such as novel-view synthesis, 3D reconstruction, novel-view acoustic synthesis, human body and face analysis, and training generative models. We provide a benchmark for training and evaluating novel-view synthesis, with two scenarios of different difficulty. Finally, we evaluate several baseline state-of-the-art methods on the new benchmark.

Arbitration Agreements in Commercial Aviation Contracts and the Hague Court of Arbitration for Aviation

Ned Beale, Ben Graham-Evans

Political Science and International Relations · Air and Space Law · · cited 1× · DOI ↗ · more like this ·

An overview of the use of arbitration agreements in commercial aviation contracts, especially lease agreements, with particular reference to the establishment of the Hague Court of Arbitration for Aviation (‘Hague CAA’) in July 2022. It considers the advantages and disadvantages of arbitration agreements, as an alternative to state court jurisdiction clauses, in commercial aviation contracts, surveys the current provision of arbitration services to the aviation sector, examines the service offered by the Hague CAA and analyses the potential benefits and risks of writing Hague CAA arbitration agreements into new leases and contracts. Aviation, arbitration, court, enforcement, Hague, insurance, jurisdiction

Tariffs (Tariff)

Ben Graham-Evans

Political Science and International Relations · Edward Elgar Publishing eBooks · · cited 0× · DOI ↗ · more like this ·

The Elgar Concise Encyclopedia of Aviation Law is an exemplary reference book providing a comprehensive overview of the origins, current landscape, and defining characteristics of the dynamic field of aviation law.

Cut-Through Clauses in Aviation Insurance and Reinsurance Policies

Ned Beale, Ben Graham-Evans

Political Science and International Relations · Air and Space Law · · cited 0× · DOI ↗ · more like this ·

This article analyses the operation of cut-through clauses in aviation insurance and reinsurance policies, with particular reference to the Lloyd’s Market Association (LMA) standard form cutthrough endorsement, AVN 109. It does so from the perspectives of the usual parties to aviation insurance policies, including on the insured’s side owners, operators, lessors and financiers, and on the underwriters’ side insurers and reinsurers. It considers cut-through clauses’ operation in the context of fronting arrangements, insurer insolvency (which brings into play the anti-deprivation principle), other bars on payment and coverage disputes. An insured’s ability to compel payment directly from the relevant insurer and the legal proceedings that may be required are discussed. The analysis is provided primarily under English law, including reference to the Contracts (Rights of Third Parties) Act…

Logarithmically larger deletion codes of all distances

Noga Alon, Gabriela Bourla, Ben Graham +2 more

math.CO · · arXiv ↗ · more like this ·

The deletion distance between two binary words $u,v \in \{0,1\}^n$ is the smallest $k$ such that $u$ and $v$ share a common subsequence of length $n-k$. A set $C$ of binary words of length $n$ is called a $k$-deletion code if every pair of distinct words in $C$ has deletion distance greater than $k$. In 1965, Levenshtein initiated the study of deletion codes by showing that, for $k\ge 1$ fixed and $n$ going to infinity, a $k$-deletion code $C\subseteq \{0,1\}^n$ of maximum size satisfies $Ω_k(2^n/n^{2k}) \leq |C| \leq O_k( 2^n/n^k)$. We make the first asymptotic improvement to these bounds by showing that there exist $k$-deletion codes with size at least $Ω_k(2^n \log n/n^{2k})$. Our proof is inspired by Jiang and Vardy's improvement to the classical Gilbert--Varshamov bounds. We also establish several related results on the number of longest common subsequences and shortest common…