203 research outputs found
Towards Post Development in India. Lessons of Community Resilience in Times of Crises
No full textAuthor: Ashish Kothari is an Indian environmentalist working on development – environment interface, biodiversity policy, and alternatives. Environment and wildlife have been his passions since high school in Delhi, when (in 1978-79) he helped found Kalpavriksh, a non-profit organisation in India which deals with environmental and development issues. Since then, he has been associated with peoples’ movements like Narmada Bachao Andolan and Beej Bachao Andolan. He is also involved in coordinateing national and global networks like Vikalp Sangam and Global Tapestry of Alternatives
Localization and instability in sheared granular materials: role of friction and vibration
No full textShear banding and stick-slip instabilities have been long observed in sheared granular materials. Yet, their microscopic underpinnings, interdependencies and variability under different loading conditions have not been fully explored. Here, we use a non-equilibrium thermodynamics model, the Shear Transformation Zone theory, to investigate the dynamics of strain localization and its connection to stability of sliding in sheared, dry, granular materials. We consider frictional and frictionless grains as well as presence and absence of acoustic vibrations. Our results suggest that at low and intermediate strain rates, persistent shear bands develop only in the absence of vibrations. Vibrations tend to fluidize the granular network and de-localize slip at these rates. Stick-slip is only observed for frictional grains and it is confined to the shear band. At high strain rates, stick-slip disappears and the different systems exhibit similar stress-slip response. Changing the vibration intensity, duration or time of application alters the system response and may cause long-lasting rheological changes. We analyse these observations in terms of possible transitions between rate strengthening and rate weakening response facilitated by a competition between shear induced dilation and vibration induced compaction. We discuss the implications of our results on dynamic triggering, quiescence and strength evolution in gouge filled fault zones.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2019-05-01The student, Konik Kothari, accepted the attached license on 2017-04-25 at 17:58.The student, Konik Kothari, submitted this Thesis for approval on 2017-04-25 at 18:10.This Thesis was approved for publication on 2017-04-26 at 18:28.DSpace SAF Submission Ingestion Package generated from Vireo submission #11071 on 2017-08-10 at 15:07:07Made available in DSpace on 2017-08-10T20:33:26Z (GMT). No. of bitstreams: 3
KOTHARI-THESIS-2017.pdf: 2654936 bytes, checksum: 03da604f37f0d94f5ba83b7593a10549 (MD5)
STZ1D.py: 32665 bytes, checksum: 3365b7018559d4e468c1a72fb94323b9 (MD5)
LICENSE.txt: 4210 bytes, checksum: 50c534fb605310ff3a22adc8eac2bd3b (MD5)
Previous issue date: 2017-04-26Embargo set by: Colleen Fallaw for item 102842
Lift date: 2019-08-10T21:27:21Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 102842 on 2019-08-11T09:15:17Z
Use of sequential diagnostic pain blocks in a patient of posttraumatic complex regional pain syndrome-not otherwise specified complicated by myofascial trigger points and thoracolumbar pain syndrome
No full textWe are presenting a case of posttraumatic lower limb Complex regional pain syndrome – Not otherwise specified (CRPS – NOS). As it was not treated in acute phase, the pain became chronic and got complicated by myofascial and thoracolumbar pain syndrome. This case posed us a diagnostic challenge. We used sequential diagnostic pain blocks to identify the pain generators and successfully treat the patient. We used diagnostic blocks step by step to identify and treat pain generators – T12,L1 and L2 Facet joints, Lumbar sympathetic block for CRPS NOS and Trigger point injection with dry needling for myofascial pain syndrome. This case highlights the facet that additional pain generators unrelated to original pain may complicate the presentation. Identifying these pain generators requires out of box thinking and high index of suspicion
Comparative genomic analyses of the cyanobacterium, Lyngbya aestuarii BL J, a powerful hydrogen producer.
No full textThe filamentous, non-heterocystous cyanobacterium Lyngbya aestuarii is an important contributor to marine intertidal microbial mats system worldwide. The recent isolate L. aestuarii BL J, is an unusually powerful hydrogen producer. Here we report a morphological, ultrastructural and genomic characterization of this strain to set the basis for future systems studies and applications of this organism. The filaments contain circa 17 μm wide trichomes, composed of stacked disk-like short cells (2 μm long), encased in a prominent, laminated exopolysaccharide sheath. Cellular division occurs by transversal centripetal growth of cross-walls, where several rounds of division proceed simultaneously. Filament division occurs by cell self-immolation of one or groups of cells (necridial cells) at the breakage point. Short, sheath-less, motile filaments (hormogonia) are also formed. Morphologically and phylogenetically L. aestuarii belongs to a clade of important cyanobacteria that include members of the marine Trichodesmiun and Hydrocoleum genera, as well as terrestrial Microcoleus vaginatus strains, and alkalyphilic strains of Arthrospira. A draft genome of strain BL J was compared to those of other cyanobacteria in order to ascertain some of its ecological constraints and biotechnological potential. The genome had an average GC content of 41.1 %. Of the 6.87 Mb sequenced, 6.44 Mb was present as large contigs (>10,000 bp). It contained 6515 putative protein-encoding genes, of which, 43 % encode proteins of known functional role, 26 % corresponded to proteins with domain or family assignments, 19.6 % encode conserved hypothetical proteins, and 11.3 % encode apparently unique hypothetical proteins. The strain’s genome reveals its adaptations to a life of exposure to intense solar radiation and desiccation. It likely employs the storage compounds, glycogen and cyanophycin but no polyhydroxyalkanoates, and can produce the osmolytes, trehalose and glycine betaine. According to
Collaborative Sampling in Generative Adversarial Networks
No full textThe standard practice in Generative Adversarial Networks (GANs) discards the discriminator during sampling. However, this sampling method loses valuable information learned by the discriminator regarding the data distribution. In this work, we propose a collaborative sampling scheme between the generator and the discriminator for improved data generation. Guided by the discriminator, our approach refines the generated samples through gradient-based updates at a particular layer of the generator, shifting the generator distribution closer to the real data distribution. Additionally, we present a practical discriminator shaping method that can smoothen the loss landscape provided by the discriminator for effective sample refinement. Through extensive experiments on synthetic and image datasets, we demonstrate that our proposed method can improve generated samples both quantitatively and qualitatively, offering a new degree of freedom in GAN sampling.VIT
Deep Learning Methods for Socially-Aware Human Trajectory Forecasting
No full textThe ability to forecast human motion, called ``human trajectory forecasting", is a critical requirement for mobility applications such as autonomous driving and robot navigation. Humans plan their path taking into account what might happen in the future. Similarly, the decision-making algorithm of autonomous systems should predict how their environment will evolve in the future. This thesis focuses on developing deep learning methods for forecasting human motion.
In the first part of this thesis, we tackle the fundamental challenges of social interaction modelling and multimodality. Social interactions dictate how the motion of a human is affected by others. Current deep learning methods often struggle to model these interactions between trajectory sequences. To promote interaction-awareness in forecasting models, we develop a training paradigm that explicitly focuses on samples that undergo interactions and incorporates model uncertainty. Furthermore, we build a taxonomy of existing interaction encoders and propose an optimal design that is robust to the real-world noise. In addition to modelling interactions, a good trajectory forecasting model must account for the multimodal nature of the prediction, i.e., the possibility of having multiple plausible futures given the past observations. To tackle multimodality, we present a socially-aware generative adversarial network that leverages recent advances in sequence modelling, and has the ability to model the temporal evolution of social interactions. Furthermore, we develop a collaborative sampling technique that refines the bad generated predictions at test time.
In the second part of this thesis, we focus on two challenges specific to the real-world deployment of forecasting models: interpretability and adaptability. While neural networks have the capacity to learn complex interactions, it is difficult to understand the reason behind their predictions. Thus, we develop a framework that combines the interpretability of the classical models with the predictive power of neural networks. With regards to adaptability, existing deep forecasting models suffer from inferior performance when they encounter novel scenarios. We develop a strategy to adapt a pre-trained forecasting model to a target domain using limited samples. In particular, we introduce motion style adapters that identify and adjust the target domain-specific features. Throughout this thesis, experiments on synthetic and real-world forecasting datasets validate the effectiveness of our proposed methods.VIT
Safety-Compliant Generative Adversarial Networks for Human Trajectory Forecasting
No full textHuman trajectory forecasting in crowds presents the challenges of modelling social interactions and outputting collision-free multimodal distribution. Following the success of Social Generative Adversarial Networks (SGAN), recent works propose various GAN-based designs to better model human motion in crowds. Despite superior performance in reducing distance-based metrics, current networks fail to output socially acceptable trajectories, as evidenced by high collisions in model predictions. To counter this, we introduce SGANv2: an improved safety-compliant SGAN architecture equipped with spatio-temporal interaction modelling and a transformer-based discriminator. The spatio-temporal modelling ability helps to learn the human social interactions better while the transformer-based discriminator design improves temporal sequence modelling. Additionally, SGANv2 utilizes the learned discriminator even at test-time via a collaborative sampling strategy that not only refines the colliding trajectories but also prevents mode collapse, a common phenomenon in GAN training. Through extensive experimentation on multiple real-world and synthetic datasets, we demonstrate the efficacy of SGANv2 to provide socially-compliant multimodal trajectories.VIT
- …
