Indian Institute of Science Bangalore
etd@IISc Electronic Theses and Dissertations at Indian Institute of ScienceNot a member yet
6204 research outputs found
Sort by
Deep Convolutional and Generative Networks for Ocean Synoptic Feature Extraction and Super Resolution from Remotely Sensed Images
No full textAccurate extraction of Synoptic Ocean Features and Downscaling of Ocean Features is crucial
for climate studies and the operational forecasting of ocean systems. With the advancement of
space and sensor technologies, the amount of remote-sensing ocean data is rising sharply. There
is a need for precise and reliable algorithms to extract information from such remotely sensed
datasets. Deep learning algorithms have shown significant superiority over traditional physical
or statistical methods for several remote-sensing applications. Two important applications are
ocean synoptic feature extraction (needed to extract useful information submerged in data)
and downscaling of satellite images (needed due to insufficient resolution of current imaging
sensors). This thesis introduces two novel deep learning algorithms: W-Net (for Ocean Feature
Extraction) and PF-GAN-SR (for Downscaling of Sea Surface Temperature Satellite Images).
Ocean Synoptic Feature Extraction: For operational regional models of the North Atlantic,
skilled human operators visualize and extract the Gulf Stream and Rings (Warm and
Cold Eddies) through a time-consuming manual process. There is a need for an automated
dynamics-inspired system to extract Gulf Stream and Rings. We have developed a deep learning
system (W-Net) that extracts the Gulf Stream and Rings from concurrent satellite images
of sea surface temperature (SST) and sea surface height (SSH). Our approach's novelty is
that the above extraction task is posed as a multi-label semantic image segmentation problem
solved by developing and applying a deep convolutional neural network with two parallel
Encoder-Decoder networks (one branch for SST and the other for SSH), implemented as a WNet.
W-Net is the first neural architecture and deep learning system developed for automated
synoptic ocean feature segmentation. For the Gulf Stream, we obtain 82.7% raw test accuracy
and a low error of 4.39% in the detected path length. For the Rings, we obtain more than 71%
raw eddy detection accuracy.
Downscaling of Sea Surface Temperature Satellite Images: The unavailability of
high-resolution remotely sensed images affects the quality of ocean forecasting and ocean feature extraction. Typical downscaling for geophysical applications is achieved using bi-linear/ bicubic
interpolation, which is not good for large downscaling ratios. To improve the current
state of the art, we developed a Bayesian algorithm for Super-Resolution (Downscaling) of
lower resolution geophysical fields observed by satellites. The key novelty is the development
and use of Generative Adversarial Networks (GAN) to learn the prior probability distribution of
the high-resolution geophysical fields from historical data and/or model forecasts. The trained
GAN is used to sample from the high-resolution prior and a Particle Filter along with the low resolution
data (observation) is used to obtain the posterior high-resolution geophysical field.
The resultant algorithm has been named the Particle Filter Generative Adversarial Network
super-resolution (PF-GAN-SR) algorithm. PF-GAN-SR is applied to downscale sea surface
temperature fields in the northwest Atlantic Ocean. Results show consistent performance across
different downscaling ratios. Notably, the high-resolution fields obtained from PF-GAN-SR
have a better similarity score with the true high-resolution field as compared to existing Super-
Resolution methods
Empirical Studies of Load Generation for Data Center Networks
No full textData center networks form the backbone for modern Internet giants like Google, Facebook, Microsoft, and Amazon. Such networks consist of a layer of a large number of storage and compute servers connected through two or three layers of switches. Depending on the target application, data center networks can be designed to have the desired properties. In many cases, the service providers have to meet service level agreements for target applications. These
agreements include the provision for redundant paths, the acceptable level of encryption, and the acceptable latency. The design choice available at the onset is architecture selection. Once the architecture is fixed, the general design choices include path selection, admission control, and encryption levels.
The performance of a data center network depends critically on the traffic profile in the network. Unfortunately, most traffic data is proprietary and seldom available to academic researchers. The industry benchmark performance tests are performed for periodic and burst packet traffic transmission. These are not representative of real traffic in the data center networks. In this work, we focus on the problem of representative load generation for the
empirical studies of data center networks through emulation and simulation.
The main characteristic of traffic is the inter-arrival time of flows for each source destination pair, their corresponding lifetimes, and their bandwidths. In this thesis, we study the type of traffic required and how to generate realistic traffic that is rich enough to test the data center networks. We propose an open source simulation framework to understand the type of traffic required and an open source realistic traffic generator which plays real life traffic traces, which
would help us understand the response of the data center networks in production. The traffic generator contains a library of traffic patterns that can be used for the load testing by varying lifetime, packet types, data rate and a few other traits obtained from traffic archive groups like MAWI, NLANR, WIDE, etc
Extraction of Equivalent Uniaxial Plastic and Viscoplastic Behavior from Bending Using a Mechanistic Approach
No full textThe present work is aimed at the extraction of material’s yield and creep parameters using
cantilevers by conducting constant deflection rate and constant load tests respectively.
Under the assumption of Euler-Bernoulli’s beam theory, stress and strain components
are considered along beam length only. A stress and strain gradient throughout the
cantilever makes bending rich in information as stress-strain rate-strain at every location
of the beam corresponds to a single uniaxial test. By extracting these stress-strain rate strain
information from multiple locations of the cantilever throughout deformation, high
throughput feature of bending is utilized by extracting all the material yield and creep
parameters from a single cantilever. The position dependent strain is measured using
digital image correlation (DIC) in this work. The estimation of stress in a cantilever
during non-linear deformation (i.e., non-linear dependence of stress on strain and/or strain
rate), however, needs numerical methods which solve for stress distribution throughout
the cantilever using equilibrium equations based on material’s constitutive behavior, beam
geometry, loading and boundary conditions. Such numerical methods are developed to
gain understanding of the stress redistribution which is found to be transient in nature
and finally saturates when permanent (plastic or creep) strain are large (∼3-4%) such
that elastic strain are negligible. The influence of material’s yield and creep parameters
on stress redistribution profile and associated timelines is studied and limitations of the
existing analytical expressions for saturated stress profile are discussed. The numerical
method is also utilized to estimate the effect of limitations on strain accuracy measured
by DIC on the extracted parameters. Therefore, the present work aims at developing
cantilever bending combined with DIC for strain measurement as an alternative testing
technique to extract material’s yield and creep parameters from a single cantilever.
The procedure for extraction of yield parameters like yield strength and strain hardening
exponent is established for a beam of tension-compression symmetric, strain rateinsensitive
material using the concept of the ‘invariant point’. These points are identified
using numerical methods as the unique locations at every cross-section of the cantilever,
one each in tensile and compressive regions, where stress remains almost unchanged during
stress redistribution. The hardening exponent and yield strength are measured from a single
cantilever with better statistics and thereby improved reliability using strain measured
at the invariant points in tensile and compressive regions within an accuracy of 99.5% and
90% respectively for pure copper and aluminium alloys. The procedures for measuring
strain rate sensitivity for a tension-compression symmetric, strain-rate sensitive material
and yield parameters for a tension-compression asymmetric, strain-rate insensitive material
are also proposed and the challenges are highlighted with the understanding that
bending has the potential to measure yield parameters for these latter systems as well in
future.
The timelines associated with stress saturation under creep deformation have been
quantified using numerical methods in terms of a parameter stress saturation time (SST).
Based on the recommendations obtained from SST, loading profile for T22 boiler steel
is redesigned in the form of small loading steps due to which stress gets sufficient time
to relax during loading itself and does not exceed yield strength during redistribution.
Thus, creep parameters can be extracted at loads as high as yield strength, which is not
possible otherwise. This makes testing faster and thereby efficient because creep rates are
higher at high load and steady state is achieved faster. In contrast, a high SST at low
loads has been identified to explain the misinterpretation of experimental data in terms
of mechanism shift at low loads for P91 steel.
The numerical method is further developed to include primary creep response at loads
above yield which holds relevance to room-temperature creep response of an hcp system,
i.e., titanium alloy Ti-6Al. In case of Ti-6Al, it is found from uniaxial creep tests conducted above yield that prior plastic deformation does not affect creep behaviour which
implies that plasticity affects only the initial stress distribution. The invariant points
which remain invariant to stress redistribution even under the combined effect of creep
and plastic deformation are identified based on the numerical methods. The strain at
these locations in tension as well as in compression, measured using DIC, are utilized to
extract equivalent primary creep response for Ti-6Al using a single cantilever. Therefore,
the present work aims at establishing bending as the testing technique to measure yield
and creep parameters for a range of materials (FCC, BCC, HCP) and testing conditions
(RT-600oC) utilizing minimum sample volume with reduced testing and better statistics.ARDB 0242, IMPRINT 000
Landmark Estimation and Image Synthesis Guidance using Self-Supervised Networks
No full textThe exponential rise in the availability of data over the past decade has fuelled research in deep learning. While supervised deep learning models achieve near-human performance using annotated data, it comes with an additional cost of annotation. Additionally, there could be ambiguity in annotations due to human error. While an image classification task assigns one label to the whole image, as we increase the granularity of the task to landmark estimation, the annotator needs to pinpoint the landmark accurately. The self-supervised learning (SSL) paradigm overcomes these concerns by using pretext task based objectives to learn from large-scale unannotated data. In this work, we show how to extract relevant signals from pretrained self-supervised networks for a) a discriminative task of landmark estimation under limited annotations, and b) increasing perceptual quality of the images generated by generative adversarial network.
In this first part, we demonstrate the emergent correspondence tracking properties in the non-contrastive SSL framework. Using this as supervision, we propose LEAD which is an approach to discover landmarks from an unannotated collection of category-specific images. Existing works in self-supervised landmark detection are based on learning dense (pixel-level) feature representations from an image, which are further used to learn landmarks in a semi-supervised manner. While there have been advances in self-supervised learning of image features for instance-level tasks like classification, these methods do not ensure dense equivariant representations. The property of equivariance is of interest for dense prediction tasks like landmark estimation. In this work, we introduce an approach to enhance the learning of dense equivariant representations in a self-supervised fashion. We follow a two-stage training approach: first, we train a network using the BYOL objective which operates at an instance level. The correspondences obtained through this network are further used to train a dense and compact representation of the image using a lightweight network. We show that having such a prior in the feature extractor helps in landmark detection, even under a drastically limited number of annotations while also improving generalization across scale variations.
Next, we utilize the rich feature space from the SSL framework as a “naturalness” prior to alleviate unnatural image generation from Generative Adversarial Networks (GAN), which is a popular class of generative models. Progress in GANs has enabled the generation of high-resolution photorealistic images of astonishing quality. StyleGANs allow for compelling attribute modification on such images via mathematical operations on the latent style vectors in the W/W+ space that effectively modulates the rich hierarchical representations of the generator. Such operations have recently been generalized beyond mere attribute swapping in the original StyleGAN paper to include interpolations. In spite of many significant improvements in StyleGANs, they are still seen to generate unnatural images. The quality of the generated images is a function of, (a) richness of the hierarchical representations learned by the generator, and, (b) linearity and smoothness of the style spaces. In this work, we propose Hierarchical Semantic Regularizer (HSR) which aligns the hierarchical representations learnt by the generator to corresponding powerful features learned by pretrained networks on large amounts of data. HSR not only improves generator representations but also the linearity and smoothness of the latent style spaces, leading to the generation of more natural-looking style-edited images. To demonstrate improved linearity, we propose a novel metric - Attribute Linearity Score (ALS). A significant reduction in the generation of unnatural images is corroborated by improvement in the Perceptual Path Length (PPL) metric by 15% across different standard datasets while simultaneously improving the linearity of attribute-change in the attribute editing tasks
Development of Solution Methodologies for Berth Allocation Problem with Discrete Berth Layout (DBL), Dynamic Arrival of Vessels (DAV), Deterministic Handling Time (DHT) of Vessels, Vessel Time Windows (VTW) and Vessel Eligibility Criteria (VEC)
No full textThe Container Terminal (CT) act as an important node in a supply chain industry that connects water
and land for transportation of the containers carrying goods. A CT facilitates a vessel by allocating a
berth for unloading and loading of the containers. During this process, the unloaded/loaded containers
are stored/retrieved in/from the yard area. The total time taken for various activities for serving a vessel
at a CT is an important performance indicator that drives competitiveness among container terminals.
Therefore, for enhancing the performance of a CT, a CT is continuously trying to optimize various
decisions that are expected to control various activities associated with servicing a vessel at a CT. One
of the important operational decisions associated with servicing a vessel is the berth allocation decision,
which serves as an important input to various other management decisions at CT and has a major
influence on the total service time of vessel at the CT.
In this study, the important decision problem: a berth allocation problem (BAP) is addressed as an
independent decision problem. Based on the closely related literature review on BAP, this study defines
a new research problem on BAP, considering simultaneously the problem characteristics of BAP such
as Discrete Berth Layout (DBL), Dynamic Arrival of Vessel (DAV), Deterministic Handling Time
(DHT) of vessel, Vessel Time Windows (VTW), and Vessel Eligibility Criteria (VEC). This newly
proposed research problem is referred as BAP with DBL-DAV-DHT-VTW-VEC in this study. The
main objective for the newly defined BAP is to simultaneously minimize the sum of the cost (called as
minimizing the Net Total Cost (NTC): Handling Cost of Vessel (HCV), Penalty Cost of Waiting of
Vessel (PCWV), Penalty Cost due to Late Departure of Vessel (PCLDV) and Benefits due to Early
Departure of Vessels (BEDV)). Subsequently, this study tries to incorporate losses due to idle time of
berth by including Penalty Cost due to Idle Time of Berth (PCITB) in NTC. In addition, to incorporate
unequal priorities to the vessels, this study considers Total Weighted Cost (TWC) as an objective.
Furthermore, this study deals with three due-date based objectives: such as minimizing the Number of
Tardy/late Vessels (NTV), maximizing the On Time Delivery (OTD) rate of vessels, and minimizing
the maximum lateness (ML), applied one at a time, as an objective for the new BAP considered in this
study
Communication Overlapping Krylov Subspace Methods for Distributed Memory Systems
No full textMany high performance computing applications in computational fluid dynamics, electromagnetics etc. need to solve a linear system of equations . For linear systems where is generally large and sparse, Krylov Subspace methods (KSMs) are used. In this thesis, we propose communication overlapping KSMs. We start with the Conjugate Gradient (CG) method, which is used when is sparse symmetric positive definite. Recent variants of CG include a Pipelined CG (PIPECG) method which overlaps the allreduce in CG with independent computations i.e., one Preconditioner (PC) and one Sparse Matrix Vector Product (SPMV).
As we move towards the exascale era, the time for global synchronization and communication in allreduce increases with the large number of cores available in the exascale systems, and the allreduce time becomes the performance bottleneck which leads to poor scalability of CG. Therefore, it becomes necessary to reduce the number of allreduces in CG and adequately overlap the larger allreduce time with more independent computations than the independent computations provided by PIPECG. Towards this goal, we have developed PIPECG-OATI (PIPECG-One Allreduce per Two Iterations) which reduces the number of allreduces from three per iteration to one per two iterations and overlaps it with two PCs and two SPMVs. For better scalability with more overlapping, we also developed the Pipelined s-step CG method which reduces the number of allreduces to one per s iterations and overlaps it with s PCs and s SPMVs. We compared our methods with state-of-art CG variants on a variety of platforms and demonstrated that our method gives 2.15x - 3x speedup over the existing methods.
We have also generalized our research with parallelization of CG on multi-node CPU systems in two dimensions. Firstly, we have developed communication overlapping variants of KSMs other than CG, including Conjugate Residual (CR), Minimum Residual (MINRES) and BiConjugate Gradient Stabilised (BiCGStab) methods for matrices with different properties. The pipelined variants give up to 1.9x, 2.5x and 2x speedup over the state-of-the-art MINRES, CR and BiCGStab methods respectively. Secondly, we developed communication overlapping CG variants for GPU accelerated nodes, where we proposed and implemented three hybrid CPU-GPU execution strategies for the PIPECG method. The first two strategies achieve task parallelism and the last method achieves data parallelism. Our experiments on GPUs showed that our methods give 1.45x - 3x average speedup over existing CPU and GPU-based implementations. The third method gives up to 6.8x speedup for problems that cannot be fit in GPU memory. We also implemented GPU related optimizations for the PIPECG-OATI method and show performance improvements over other GPU implementations of PCG and PIPECG on multiple nodes with multiple GPUs
Design and Development of a Semi-automated System for Electro-thermo-mechanical (Etm) Phenotyping of Breast Tissues Ex-vivo
No full textBreast cancer currently accounts for 25% of all cancers diagnosed in women globally. The conventional confirmatory diagnosis of breast cancer involves histological analysis using hematoxylin and eosin (H&E) staining, followed by immunohistochemical analysis for breast cancer biomarkers. For surgical margin assessment within the operating room (OR), the standard technique is frozen section examination. However, this process involves sending the biopsy tissue out of the OR and into the pathology laboratories, with the analysis time for each sample ranging between 30 min and 2 h. This adds to the time of diagnosis as well as the surgery. Therefore, tools and technologies that can provide a rapid and label-free assessment of breast biopsies to delineate between benign, malignant, and adjacent normal tissue can be a significant adjunct to routine diagnostics. This thesis proposes a label-free multimodal approach for breast tumor delineation using electro-thermo-mechanical (ETM) phenotyping with a MEMS-based semi-automated system.
First, the design and development of the RapidET system, the MEMS-based semi-automated platform integrated with microchips and electronic modules for biophysical characterization of tissues, is presented. Microchips for electro-thermal characterization of the samples were first integrated with the system. The microchips, fabricated on a silicon substrate, incorporate a platinum microheater, interdigitated electrodes (IDEs), and resistance temperature detectors (RTDs) as on-chip sensing elements. The measurement accuracy of the system was first validated on a murine xenograft tumor model through electro-thermal characterization of ex vivo tumors and healthy tissues. Next, the bulk resistivity (ρB), surface resistivity (ρS), and thermal conductivity (k) of deparaffinized and formalin-fixed paired tumor and adjacent normal breast biopsy samples from N = 8 subjects were measured. The bulk and surface resistivity of tumors showed a significant increase with temperature compared to the adjacent normal. The tumor tissues were also observed to have a significantly lower (0.309 ± 0.02 Wm-1K-1) thermal conductivity than normal (0.563 ± 0.028 Wm-1K-1).
Next, the scaled increase in resistivity with temperature observed for the tumor tissues was further explored by performing a bimodal temperature and frequency-dependent electrical transport characterization of the samples (N = 10). Temperature-dependent direct current (DC) transport was modeled under the realm of general effective medium theory. Critical temperature (Tc) as a model fit parameter was found to be higher for adjacent normal (42.8 ± 2.0 ºC) compared to the tumor (36.5 ± 0.9 ºC), indicating an early transition from conducting to the insulating regime in tumor tissues. Frequency-dependent alternating current (AC) transport was observed to follow the scaling law, which is used to model disordered systems and growth in biological systems. The parameter onset frequency fc was higher for adjacent normal (1.1 ± 0.37 MHz) than the tumor (33.5 ± 14.9 kHz), indicating higher disorder in tumor samples. The utility of the model fit parameters in classifying samples as tumor and normal was demonstrated using a support vector machine (SVM) classifier, which showed 91.7% accuracy when compared to 70% as obtained for the raw data.
Finally, the system was augmented with a microforce sensor to create a system-of-biochips (SoB) to perform electro-thermo-mechanical phenotyping. The SoB measures the electrical impedance (Z), thermal conductivity (K), mechanical stiffness (k), and viscoelastic stress relaxation (%R) of the samples. Multimodal ETM characterization performed on formalin-fixed breast biopsy samples from N = 14 subjects was able to differentiate between invasive ductal carcinoma (malignant), fibroadenoma (benign), and adjacent normal (healthy) tissues with a root mean square error of 0.2419 using a gaussian process classifier. After validation with fresh tissues, this multimodal methodology could potentially be used for delineating benign, malignant, and adjacent normal breast tissues during surgery and in pathology labs
B, N containing π- conjugated systems: room temperature phosphorescence, fullerene complexation, and cell differentiation
No full textAdvances in heteroatom (B, N, and B-N) incorporated π-conjugated systems have attracted considerable attention because of their desirable properties and widespread applications. Substitution of isoelectronic B–N units for C=C units in π-conjugated systems produce novel materials with structural similarities to all-carbon frameworks but with fundamentally altered electronic and optical properties. However, despite significant progress in this field, there is a lack of systematic studies of their fundamental properties like solid/condensed state optical behaviour of BN compounds. The objective of this thesis is to synthesize boron-nitrogen-containing compounds and to investigate their photophysical properties and applications. As part of this program, we have demonstrated the host-guest chemistry of B-N doped polyaromatic heterocycles (BN-PAHCs) for the first time. Further, we developed doubly B-N fused Ni(II) porphyrins with ruffled core structures and studied their opto-electrochemical behaviours. We have also explored the potential of aminoboranes as a new class of persistent room-temperature phosphorescent materials by modulating their electronic properties upon judiciously varying the substituents in the B-N unit. Novel anthracene-boron-based luminophores were synthesized. These compounds exhibit intriguing optical properties such as aggregation-induced emission enhancement (AIEE) and microviscosity-dependent emission. The unique optical features of these compounds have been successfully exploited for differentiating normal, autophagic, and apoptotic cells. Encouraged by these promising results, a series of molecules consisting of both tri and tetracoordinate boron centers were synthesized. The non-planar geometry of tetracoordinate boron combined with sterically demanding tricoordinate dimesitylboron unit inhibits π-stacking interactions and leads to strong luminescence in the solid state. Furthermore, we accidentally discovered a rare class of tribophosphorescent materials with molecular architecture composed of triboluminescent moiety (anthracene) and an ISC facilitator (-Bpin). This serendipitous discovery unveiled a new design strategy for developing tribophosphorescent molecules. This thesis contains all of these fascinating findings.Indian Institute of Science, Bangalor
Elucidating the Role of Cyclic GMP in Diarrhoea and Intestinal Inflammation
No full textCyclic guanosine 3’,5’-monophosphate (cGMP) performs a wide range of functions in various cell types and tissues. The cellular levels of cGMP are maintained by the enzymatic conversion of guanosine 5’-triphosphate (GTP) to cGMP by guanylyl cyclases, which can also be membrane-associated receptors. Receptor guanylyl cyclase C (GC-C; gene GUCY2C) is predominantly expressed on the apical surface of the intestinal epithelial cells. GC-C is activated by peptide hormones guanylin and uroguanylin and the heat-stable enterotoxin (ST) produced by enterotoxigenic E.coli that causes traveller’s diarrhoea.
Several disease-causing mutations in GUCY2C have been reported. Patients with gain-of-function mutations show diarrhoea and inflammatory bowel disease (IBD). Several cases of paediatric IBD have also been associated with mutations in GUCY2C. This study addresses the physiological implications of increased cGMP using a transgenic mouse model harbouring the first-identified hyperactive mutation in human patients leading to familial GUCY2C diarrhoea syndrome (FGDS). Mice with hyperactive GC-C showed increased levels of cGMP in intestinal epithelial cells, which led to activation of Cftr and inhibition of Nhe3, resulting in diarrhoea-like symptoms and increased luminal pH and faecal sodium levels. Global transcriptome analysis of the distal colon revealed activation of the interferon signalling pathway, and transgenic mice showed greater susceptibility to DSS-induced colitis.
Histological analysis of the terminal ileum revealed a reduction in functional Paneth cells, goblet cells, and mucus barrier. The barrier integrity of the small intestine was compromised in these mice. Global transcriptome analysis of the terminal ileum revealed a Th1-type gene signature. Immune cell profiling across the gut-associated lymphoid tissue (GALT) showed reduced regulatory dendritic cells and an increased abundance of CD4+ Th cells. Increased levels of Stat1 were observed in the ileal epithelial cells of these mice, along with elevated expression of interferon-stimulated genes.
Small intestinal organoids were prepared from wild type and transgenic mice. The organoids from transgenic mice showed greater swelling in the presence of ST and uroguanylin due to increased fluid secretion into the lumen. Administration of cGMP increased Stat1 phosphorylation in the intestinal organoids. Our observations suggest that high cGMP levels have epithelial cell-extrinsic and cell-intrinsic roles in inducing intestinal inflammation. Furthermore, the administration of zinc inhibited the activity of GC-C and reduced diarrhoea and intestinal inflammation in the transgenic mice. Thus, the similarities observed in these transgenic mice with that of chronic diarrhoea and IBD patients indicate that they can be used as a pre-clinical model to understand the effects of chronically elevated cGMP on intestinal pathophysiology and for identifying novel therapeutic strategies for patients with hyperactivating mutations in GUCY2C
Enantioselective Annulation Reactions: From Fischer Indolization to de novo Arene Construction
No full textIn summary, we have developed the first decarboxylative [4+2]-annulation of ethynyl
benzoxazinanones with azlactones for the enantioselective synthesis of cyclic α-quaternary
α-acylaminoamides.25 This direct and modular approach combines dipolar copper-allenylidene
intermediates and enolates generated from azlactones under bifunctional tertiary aminourea
catalysis in a cooperative fashion. The resulting 3,4-dihydroquinolin-2-ones 3.43, bearing vicinal
tertiary and quaternary stereogenic centers, are formed as a single diastereomer with good to
excellent enantioselectivity. The products are densely functionalized and the synthetic versatility of
the terminal alkyne group is demonstrated through a number of synthetic elaborations. This reaction
represents the first direct catalytic enantioselective route to cyclic α-quaternary α-acylaminoamide