More Like this

1. Developing a Smart Façade System Controller for Wind-Induced Vibration Mitigation in Tall Buildings

   https://doi.org/10.1115/SMASIS2019-5674  

   Abdelaziz, Khalid M.; Hobeck, Jared D. (December 2019, Smart Materials, Adaptive Structures and Intelligent Systems)

   null (Ed.)

   Tall and slender buildings often endure disturbances resulting from winds composed of various mean and fluctuating velocities. These disturbances result in discomfort for the occupants as well as accelerated fatigue life cycles and premature fatigue failures in the building. This work presents the development of a smart morphing façade (Smorphaçade) system that dynamically alters a buildings’ external shape or texture to minimize the effect of wind-induced vibrations on the building. The Smorphacade system is represented in this work by a series of plates that vary their orientation by means of a central controller module. To validate the simulation, a simple NACA0012 airfoil is simulated in a stream of air at a Reynolds number (RE) of 2 million. The pressure and viscous force profiles are captured to plot the variation of the lift force for different angles of attack that are then validated using published experimental airfoil data. After validation, the airfoil is attached to a linear spring-damper combination and is allowed to translate vertically without rotation according to the force profile captured from the surrounding air stream. A PID controller is developed to equilibrate the vertical position of the airfoil by altering its angle of attack. The model and its utility functions are implemented as an OpenFOAM® module (MSLSolid). Thereafter, the model is expanded to handle a planar case of a building floor carrying 4 controllable plates. The forces on the building profile are summed at the centroid of the building and the windward rigid body motion of the floor is estimated by reflecting the horizontal force component on a Finite Element (FE) model of the building. The time series information of the force acting on the building and the resulting oscillations are captured for exhaustive combinations of the plate angles. This data is used to build a lookup table that gives the best plate configuration for a given wind condition. A controller operates in real-time by searching the lookup table using readings of the wind condition. Preliminary results show a 94% reduction in the amplitudes of wind-induced vibrations. 

   more » « less
2. On Locally Decodable Codes in Resource Bounded Channels

   https://doi.org/10.4230/LIPIcs.ITC.2020.16  

   Blocki, J; Kulkarni, S; Zhou, S. (January 2020, Leibniz international proceedings in informatics)

   Yael Tauman Kalai and Adam D. Smith and Daniel Wichs (Ed.)

   Constructions of locally decodable codes (LDCs) have one of two undesirable properties: low rate or high locality (polynomial in the length of the message). In settings where the encoder/decoder have already exchanged cryptographic keys and the channel is a probabilistic polynomial time (PPT) algorithm, it is possible to circumvent these barriers and design LDCs with constant rate and small locality. However, the assumption that the encoder/decoder have exchanged cryptographic keys is often prohibitive. We thus consider the problem of designing explicit and efficient LDCs in settings where the channel is slightly more constrained than the encoder/decoder with respect to some resource e.g., space or (sequential) time. Given an explicit function f that the channel cannot compute, we show how the encoder can transmit a random secret key to the local decoder using f(⋅) and a random oracle 𝖧(⋅). We then bootstrap the private key LDC construction of Ostrovsky, Pandey and Sahai (ICALP, 2007), thereby answering an open question posed by Guruswami and Smith (FOCS 2010) of whether such bootstrapping techniques are applicable to LDCs in channel models weaker than just PPT algorithms. Specifically, in the random oracle model we show how to construct explicit constant rate LDCs with locality of polylog in the security parameter against various resource constrained channels. 

   more » « less
3. Neural Syntactic Preordering for Controlled Paraphrase Generation

   https://doi.org/10.18653/v1/2020.acl-main.22  

   Goyal, Tanya; Durrett, Greg (January 2020, Annual Meeting of the Association for Computational Linguistics)

   Paraphrasing natural language sentences is a multifaceted process: it might involve replacing individual words or short phrases, local rearrangement of content, or high-level restructuring like topicalization or passivization. Past approaches struggle to cover this space of paraphrase possibilities in an interpretable manner. Our work, inspired by pre-ordering literature in machine translation, uses syntactic transformations to softly "reorder" the source sentence and guide our neural paraphrasing model. First, given an input sentence, we derive a set of feasible syntactic rearrangements using an encoder-decoder model. This model operates over a partially lexical, partially syntactic view of the sentence and can reorder big chunks. Next, we use each proposed rearrangement to produce a sequence of position embeddings, which encourages our final encoder-decoder paraphrase model to attend to the source words in a particular order. Our evaluation, both automatic and human, shows that the proposed system retains the quality of the baseline approaches while giving a substantial increase in the diversity of the generated paraphrases. 

   more » « less
4. Optimal Strategic Quantizer Design via Dynamic Programming

   https://doi.org/10.1109/DCC52660.2022.00025  

   Anand, Anju; Akyol, Emrah (March 2022, 2022 Data Compression Conference (DCC))

   This paper is concerned with the quantization setting where the encoder and the decoder have misaligned objectives. We first motivate the problem via a toy example which demonstrates the intricacies of the strategic quantization problem, specifically shows that iterative optimization of the decoder and the encoder mappings may not converge to a local optimum. As a remedy, we propose a dynamic programming based optimal optimization method, inspired by the early works in the quantization theory. We then extend our approach to variable-rate (entropy-coded) quantization. We finally present numerical results obtained via the proposed algorithms. 

   more » « less
5. Multifidelity data fusion in convolutional encoder/decoder assembly networks for computational fluid dynamics

   https://doi.org/10.2514/6.2022-0803  

   Partin, Lauren; Rushdi, Ahmad A.; Schiavazzi, Daniele E. (January 2022, AIAA SCITECH 2022 Forum)

   We analyze the regression accuracy of convolutional neural networks assembled from encoders, decoders and skip connections and trained with multifidelity data. These networks benefit from a significant reduction in the number of trainable parameters with respect to an equivalent fully connected network. These architectures are also versatile with respect to the input and output dimensionality. For example, encoder-decoder, decoder-encoder or decoder-encoder-decoder architectures are well suited to learn mappings between input and outputs of any dimensionality. We demonstrate the accuracy produced by such architectures when trained on a few high-fidelity and many low-fidelity data generated from models ranging from one-dimensional functions to Poisson equation solvers in two-dimensions. We finally discuss a number of implementation choices that improve the reliability of the uncertainty estimates generated by a dropblock regularizer, and compare uncertainty estimates among low-, high- and multi-fidelity approaches. 

   more » « less