Seminar on Topics in Integrated Systems

Lecturer (assistant)
TypeSeminar
Duration3 SWS
TermWintersemester 2020/21
Language of instructionEnglish

Dates

Admission information

See TUMonline
Note: Registration (via TUMonline from 12.10.2020 to 10.11.2020) is required. Limited number of participants! Students have to choose a seminar topic before the introduction lesson. Therefore you need to contact the supervisor of the topic you are interested in. Topics are selected on a first come first served basis. Topics will be published on 9th October 2020 <a href="https://www.ei.tum.de/en/lis/teaching/seminars/seminar-on-topics-in-integrated-system-design/"> https://www.ei.tum.de/en/lis/teaching/seminars/seminar-on-topics-in-integrated-system-design/</a>

Objectives

At the end of the seminar, the students is able to present a state-of-the-art literature review in the area of integrated systems building blocks and architectures in an understandable and convincing manner. The following competencies will be acquired: * The student is able to independently analyze state-of-the-art concepts in the field of integrated systems. * The student is able to present a topic in a structured way according to problem formulation, state of the art, goals, methods, and results. * The student can present a topic according to the structure given above orally, on a poster, with a set of slides, and with a written report.

Description

Specific topics in the area of integrated circuits and systems will be offered. The participants independently work on a current scientific topic, write a paper, design a poster and present their topic in a talk. In the subsequent discussion the topic will be trated in-depth.

Prerequisites

None

Teaching and learning methods

Learning method: Students elaborate a given scientific topic by themselves in coordination with the respective research assistant. Teaching method: Introductory lessons will be given by the course coordinator, further details are discussed between research assistant and student on an individual basis. Presentation skills will be educated by a professional teacher.

Examination

Examination with the following elements: - paper of 4 pages in IEEE format - presentation of a poster - talk of 20 Minutes and subsequent questions

Recommended literature

A set of topics and related literature is given at the start of the course. Each participant selects his/her topic.

Links

Offered Topics

Seminars

[MSCE] Comparing strategies to reduce data-triffic on distributed memory architectures

[MSCE] A Survey on Spiking Neural Networks and Neuromorphic Computing

[MSCE] Efficient Training Schemes for Binary Neural Networks

[MSCE] Understanding Adversarial Robustness in the Context of 3D-Object Detection

[MSCE] In-Train Quantization and Pruning Methods for Convolutional Neural Networks

[MSCE] Hardware Architecture Search for Inference of Convolutional Neural Networks

[MSCE] Flexible On-Chip Interconnect for Runtime Reconfigurable Inference Dataflows of Neural Networks

[MSCE] Optimized CUDA Kernels for Efficient Processing of Binary and Sparse Neural Networks

Ruleset reduction algorithms in XCS classifier system.

[MSCE] A Survey on Inter-Process Communication

[MSCE] Survey of Application Profiling Tools in Linux

Survey on Multiobjective Genetic Algorithms

Feature selection for LSTM neural networks

Markov Decision Processes as Basis for Multistep Learning with LCS

LCS for Real-Time Systems

Load Balancing Techniques deployed in Network Interface Cards (NICs)

Survey on Hash Functions used for Data Mapping

Assigned Topics

Seminars

Clock recovery circuits on FPGA fabric

[MSCE] Performance and Power Models for Near-memory Computing

Update on Winter Semester 2020/21 (09.10.2020)

The seminar topics are now available. You can get in touch with the corresponding supervisors. The course information will be updated and will available for registration at TUMOnline soon.