-
Artifact Reduction in Fundus Imaging using Cycle Consistent Adversarial Neural Networks
Authors:
Sai Koushik S S,
K. G. Srinivasa
Abstract:
Fundus images are very useful in identifying various ophthalmic disorders. However, due to the presence of artifacts, the visibility of the retina is severely affected. This may result in misdiagnosis of the disorder which may lead to more complicated problems. Since deep learning is a powerful tool to extract patterns from data without much human intervention, they can be applied to image-to-imag…
▽ More
Fundus images are very useful in identifying various ophthalmic disorders. However, due to the presence of artifacts, the visibility of the retina is severely affected. This may result in misdiagnosis of the disorder which may lead to more complicated problems. Since deep learning is a powerful tool to extract patterns from data without much human intervention, they can be applied to image-to-image translation problems. An attempt has been made in this paper to automatically rectify such artifacts present in the images of the fundus. We use a CycleGAN based model which consists of residual blocks to reduce the artifacts in the images. Significant improvements are seen when compared to the existing techniques.
△ Less
Submitted 25 December, 2021;
originally announced December 2021.
-
Chip Placement with Deep Reinforcement Learning
Authors:
Azalia Mirhoseini,
Anna Goldie,
Mustafa Yazgan,
Joe Jiang,
Ebrahim Songhori,
Shen Wang,
Young-Joon Lee,
Eric Johnson,
Omkar Pathak,
Sungmin Bae,
Azade Nazi,
Jiwoo Pak,
Andy Tong,
Kavya Srinivasa,
William Hang,
Emre Tuncer,
Anand Babu,
Quoc V. Le,
James Laudon,
Richard Ho,
Roger Carpenter,
Jeff Dean
Abstract:
In this work, we present a learning-based approach to chip placement, one of the most complex and time-consuming stages of the chip design process. Unlike prior methods, our approach has the ability to learn from past experience and improve over time. In particular, as we train over a greater number of chip blocks, our method becomes better at rapidly generating optimized placements for previously…
▽ More
In this work, we present a learning-based approach to chip placement, one of the most complex and time-consuming stages of the chip design process. Unlike prior methods, our approach has the ability to learn from past experience and improve over time. In particular, as we train over a greater number of chip blocks, our method becomes better at rapidly generating optimized placements for previously unseen chip blocks. To achieve these results, we pose placement as a Reinforcement Learning (RL) problem and train an agent to place the nodes of a chip netlist onto a chip canvas. To enable our RL policy to generalize to unseen blocks, we ground representation learning in the supervised task of predicting placement quality. By designing a neural architecture that can accurately predict reward across a wide variety of netlists and their placements, we are able to generate rich feature embeddings of the input netlists. We then use this architecture as the encoder of our policy and value networks to enable transfer learning. Our objective is to minimize PPA (power, performance, and area), and we show that, in under 6 hours, our method can generate placements that are superhuman or comparable on modern accelerator netlists, whereas existing baselines require human experts in the loop and take several weeks.
△ Less
Submitted 22 April, 2020;
originally announced April 2020.
-
Early Performance Prediction of Web Services
Authors:
Ch Ram Mohan Reddy,
D. Evangelin Geetha,
K. G. Srinivasa,
T. V. Suresh Kumar,
K. Rajani Kanth
Abstract:
Web Service is an interface which implements business logic. Performance is an important quality aspect of Web services because of their distributed nature. Predicting the performance of web services during early stages of software development is significant. In this paper we model web service using Unified Modeling Language, Use Case Diagram, Sequence Diagram, Deployment Diagram. We obtain the Pe…
▽ More
Web Service is an interface which implements business logic. Performance is an important quality aspect of Web services because of their distributed nature. Predicting the performance of web services during early stages of software development is significant. In this paper we model web service using Unified Modeling Language, Use Case Diagram, Sequence Diagram, Deployment Diagram. We obtain the Performance metrics by simulating the web services model using a simulation tool Simulation of Multi-Tier Queuing Architecture. We have identified the bottle neck resources.
△ Less
Submitted 10 January, 2012;
originally announced January 2012.
-
General Methodology for develo** UML models from UI
Authors:
Ch Ram Mohan Reddy,
D. Evangelin Geetha,
K. G. Srinivasa,
T. V. Suresh Kumar,
K. Rajani Kanth
Abstract:
In recent past every discipline and every industry have their own methods of develo** products. It may be software development, mechanics, construction, psychology and so on. These demarcations work fine as long as the requirements are within one discipline. However, if the project extends over several disciplines, interfaces have to be created and coordinated between the methods of these discip…
▽ More
In recent past every discipline and every industry have their own methods of develo** products. It may be software development, mechanics, construction, psychology and so on. These demarcations work fine as long as the requirements are within one discipline. However, if the project extends over several disciplines, interfaces have to be created and coordinated between the methods of these disciplines. Performance is an important quality aspect of Web Services because of their distributed nature. Predicting the performance of web services during early stages of software development is significant. In Industry, Prototype of these applications is developed during analysis phase of Software Development Life Cycle (SDLC). However, Performance models are generated from UML models. Methodologies for predicting the performance from UML models is available. Hence, In this paper, a methodology for develo** Use Case model and Activity model from User Interface is presented. The methodology is illustrated with a case study on Amazon.com.
△ Less
Submitted 10 January, 2012;
originally announced January 2012.
