Masaki Yamada
をテンプレートにして作成
[
トップ
] [
新規
|
一覧
|
単語検索
|
最終更新
|
ヘルプ
|
ログイン
]
開始行:
[[Members-Internal]]
CENTER:SIZE(40){COLOR(blue){GT1-S1: FPGA Acceleration and Performance Study of Character Recognition with Feed-Forward Neural Network&br [順伝搬型ニューラルネットワークを用いた文字認識システムの性能調査とFPGAによる高速化]}}
*''Motivation'' [#l3b6f17e]
Feed Forward Neural Networks (FFNN) when designed to work with floating point (FP) precision performs a large number of elementary products
and sums. For each neuron of FFNN within the hidden layers, a non-linear function computation is required to determine the activation value
of the neuron. Without efficient, dedicated FP hardware, such computations can create difficulties for the whole system performance of the system, hence making the design difficult to be used in critical applications like real-time systems.
*''Goal'' [#n719c9ab]
The goal of this research is to implement a Feed Forward Neural Networks (FFNN) with floating point on FPGA. A real application, such as character recognition, should be demonstrated.
*References for Seminar [#cefa844e]
-[[Ref.1>https://drive.google.com/file/d/0B2HMlO4p7SuwZTRadm54ekg3ZG8/view?usp=sharing]]
--July 18,2017, Ch 1,2
--July %%25%%,27, 2017, Ch 3, 4
--August 4, 2017, Ch 5,6, 10:30 AM
--September 11, 2017, 10:00 AM, Character recognition implementation in Python
---COLOR(red){upload Python code and slides for FFNN}
---COLOR(red){upload Python code and slides for CNN}
-October 6, 2017, 18:00
--Convert Python code to C then to Verilog HDL
*Special Semminars [#x0bbf332]
-September 11, 2017 [[slides.pptx]]
-July 18, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQMEo6fjsQXkc27xoCQoW9nZrdMN6ApZiRT6H6Ycxg0lMIa1vuSb4dpzu29w5m_ycSP9M4VnQZdfOVP/pub?start=false&loop=false&delayms=3000]]
-July 27, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQMEo6fjsQXkc27xoCQoW9nZrdMN6ApZiRT6H6Ycxg0lMIa1vuSb4dpzu29w5m_ycSP9M4VnQZdfOVP/pub?start=false&loop=false&delayms=3000]]
-August 4, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQbryyAWDtj9S4QpPcuc1G4A3NZJS-iiEl0B-iBzvyJL7EDfft_pJ1EPONWfJCKlmtoSeHhm4lX9C_Q/pub?start=false&loop=false&delayms=3000]]
*[[NASH seminars>http://adaptive.u-aizu.ac.jp/aslint/index.php?NASH-SEMINAR]] [#tf538694]
*GT Poster [#o89a792c]
-- 2017/10/13: Midterm-poster (&ref(s1220042_midtermposter-UPD.pptx,,pptx);)
-- 2017/09/28: Midterm-poster (&ref(s1220042_midtermposter.pptx,,pptx);)
終了行:
[[Members-Internal]]
CENTER:SIZE(40){COLOR(blue){GT1-S1: FPGA Acceleration and Performance Study of Character Recognition with Feed-Forward Neural Network&br [順伝搬型ニューラルネットワークを用いた文字認識システムの性能調査とFPGAによる高速化]}}
*''Motivation'' [#l3b6f17e]
Feed Forward Neural Networks (FFNN) when designed to work with floating point (FP) precision performs a large number of elementary products
and sums. For each neuron of FFNN within the hidden layers, a non-linear function computation is required to determine the activation value
of the neuron. Without efficient, dedicated FP hardware, such computations can create difficulties for the whole system performance of the system, hence making the design difficult to be used in critical applications like real-time systems.
*''Goal'' [#n719c9ab]
The goal of this research is to implement a Feed Forward Neural Networks (FFNN) with floating point on FPGA. A real application, such as character recognition, should be demonstrated.
*References for Seminar [#cefa844e]
-[[Ref.1>https://drive.google.com/file/d/0B2HMlO4p7SuwZTRadm54ekg3ZG8/view?usp=sharing]]
--July 18,2017, Ch 1,2
--July %%25%%,27, 2017, Ch 3, 4
--August 4, 2017, Ch 5,6, 10:30 AM
--September 11, 2017, 10:00 AM, Character recognition implementation in Python
---COLOR(red){upload Python code and slides for FFNN}
---COLOR(red){upload Python code and slides for CNN}
-October 6, 2017, 18:00
--Convert Python code to C then to Verilog HDL
*Special Semminars [#x0bbf332]
-September 11, 2017 [[slides.pptx]]
-July 18, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQMEo6fjsQXkc27xoCQoW9nZrdMN6ApZiRT6H6Ycxg0lMIa1vuSb4dpzu29w5m_ycSP9M4VnQZdfOVP/pub?start=false&loop=false&delayms=3000]]
-July 27, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQMEo6fjsQXkc27xoCQoW9nZrdMN6ApZiRT6H6Ycxg0lMIa1vuSb4dpzu29w5m_ycSP9M4VnQZdfOVP/pub?start=false&loop=false&delayms=3000]]
-August 4, 2017: [[slide>https://docs.google.com/presentation/d/e/2PACX-1vQbryyAWDtj9S4QpPcuc1G4A3NZJS-iiEl0B-iBzvyJL7EDfft_pJ1EPONWfJCKlmtoSeHhm4lX9C_Q/pub?start=false&loop=false&delayms=3000]]
*[[NASH seminars>http://adaptive.u-aizu.ac.jp/aslint/index.php?NASH-SEMINAR]] [#tf538694]
*GT Poster [#o89a792c]
-- 2017/10/13: Midterm-poster (&ref(s1220042_midtermposter-UPD.pptx,,pptx);)
-- 2017/09/28: Midterm-poster (&ref(s1220042_midtermposter.pptx,,pptx);)
ページ名: