Generative AI Study Group_2ndSesssion_20230620

6.9K Views

June 21, 23

ai generative ai artificial intelligence machine learning deep learning

スライド概要

Generative Ai Study Group
2nd Session
Presentation material

今回は「Generative AI Study Group」という名前のもと、Zoomウェビナーで開催され、前回の振り返り、アンケート結果公表、名簿の作成、テーマ募集などを行いました。また、今回のセミナーにおいては、SBS情報システム土屋様により、業務効率化の実現手段としての生成系AIのセンテンスジェネレーションの紹介がありました。そして、OpenAIの多数のリリースの紹介もあり、Function callingについても解説が行われました。参加者は積極的にコメントや挙手をし、活発な交流が行われまた。

profile-image
スライド一覧

Generative Ai Study Group Master

シェア

またはPlayer版

埋め込む »CMSなどでJSが使えない場合

ダウンロード

ダウンロード(pdf - 1.26MB)

関連スライド

slide-thumbnail

Generative AI Study Group_FirstSesssion_20230606
ai generative ai artificial intelligence machine learning deep learning
user-img Kunihiro Sugiyama 5.6K
slide-thumbnail

Generative AI Study Group_振り返り会
ai generative ai machine learning deep learning artificial intelligence
user-img Kunihiro Sugiyama 5.4K
slide-thumbnail

Generative AI Study Group_11thSesssion_20231114
ai generative ai machine learning deep learning artificial intelligence
user-img Kunihiro Sugiyama 5.2K
slide-thumbnail

Generative AI Study Group_3rdSesssion_20230704
ai generative ai machine learning deep learning
user-img Kunihiro Sugiyama 5.1K
slide-thumbnail

Generative AI Study Group_4thSesssion_20230801
ai generative ai machine learning deep learning artificial intelligence
user-img Kunihiro Sugiyama 4.4K
slide-thumbnail

Generative AI Study Group_6thSesssion_20230905
ai generative ai machine learning deep learning artificial intelligence
user-img Kunihiro Sugiyama 4.4K
各ページのテキスト
1.

Generative AI Study Group

2.

Agenda https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

3.

Agenda https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

4.

Introduction • 本日の開催形式 • Zoomウェビナー • コメントや挙手はぜひ積極的にいただけると嬉しいです。 • 今回の集客について

5.

Introduction • 前回の振り返り • Web更新 • https://www.ai-tech-c.jp/generative-ai-study-group-gasg/ • アンケート結果 • 今回のアンケート • 今回からアンケート結果はWebで公表させていただきます。

6.

Introduction • 参加者名簿 • GASGの参加者名簿を作りたいです。 • 記念に、思い出に、皆様お揃いで、よろしければ。 • Form • https://forms.gle/eRmeGtfkQMkDrs7K9 • 名簿 • ROで一般公開されます。 • https://docs.google.com/spreadsheets/d/1cwA2NSgFk5EICU7TgX8WvhzrXTqzQ4ynexVdjv1MkDY/edi t?usp=sharing

7.

Introduction • テーマ募集 • テーマ募集中です!!

8.

Agenda https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

9.

Theme Presentation • SBS情報システム 土屋様 • 生成系AIのセンテンスジェネレーションを業務効率化の実現手段として検討した経緯をご紹介

10.

Agenda Next News https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

11.

Study contents • Open AI new release • 6月14日にOpenAIから多数のリリースがあった。 • LLMの性能向上 • 対応トークン長の拡大 • 値下げ • そして...

12.

Study contents Function calling

13.

Study contents • Function calling • From OpenAI API reference • In an API call, you can describe functions to gpt-3.5-turbo0613 and gpt-4-0613, and have the model intelligently choose to output a JSON object containing arguments to call those functions. The Chat Completions API does not call the function; instead, the model generates JSON that you can use to call the function in your code. https://platform.openai.com/docs/guides/gpt/function-calling

14.

Study contents • Function calling • From OpenAI API reference • In an API call, you can describe functions to gpt-3.5-turbo0613 and gpt-4-0613, and have the model intelligently choose to output a JSON object containing arguments to call those functions. The Chat Completions API does not call the function; instead, the model generates JSON that you can use to call the function in your code. https://platform.openai.com/docs/guides/gpt/function-calling

15.

Study contents • Function calling • From OpenAI API reference • In an API call, you can describe functions to gpt-3.5-turbo0613 and gpt-4-0613, and have the model intelligently choose to output a JSON object containing arguments to call those functions. The Chat Completions API does not call the function; instead, the model generates JSON that you can use to call the function in your code. https://platform.openai.com/docs/guides/gpt/function-calling

16.

Study contents User • Function calling • こうなると思ってた Business Logic LLM Service API

17.

Study contents User Business Logic LLM Service API • Function calling • 違った… 言語処理 言語処理

18.

Study contents User • Function calling • シンプルに言えば Business Logic LLM Service API

19.

Study contents • Function calling • From OpenAI API reference • The latest models (gpt-3.5-turbo-0613 and gpt-4-0613) have been fine-tuned to both detect when a function should to be called (depending on the input) and to respond with JSON that adheres to the function signature. With this capability also comes potential risks.We strongly recommend building in user confirmation flows before taking actions that impact the world on behalf of users (sending an email, posting something online, making a purchase, etc). https://platform.openai.com/docs/guides/gpt/function-calling

20.

Study contents • Function calling • From OpenAI API reference • The latest models (gpt-3.5-turbo-0613 and gpt-4-0613) have been fine-tuned to both detect when a function should to be called (depending on the input) and to respond with JSON that adheres to the function signature. With this capability also comes potential risks.We strongly recommend building in user confirmation flows before taking actions that impact the world on behalf of users (sending an email, posting something online, making a purchase, etc). https://platform.openai.com/docs/guides/gpt/function-calling

21.

Study contents • Function calling • From OpenAI API reference • The latest models (gpt-3.5-turbo-0613 and gpt-4-0613) have been fine-tuned to both detect when a function should to be called (depending on the input) and to respond with JSON that adheres to the function signature. With this capability also comes potential risks.We strongly recommend building in user confirmation flows before taking actions that impact the world on behalf of users (sending an email, posting something online, making a purchase, etc). https://platform.openai.com/docs/guides/gpt/function-calling

22.

Study contents •First impression •思ったより穏当

23.

Study contents • Fuction calling • 他にみるべき取り組みがある • Gollira • API呼び出しコードでFinetuningしたLLM • https://gigazine.net/news/20230615-gorilla/ • Langchain • Toolで作り込めば... • AutoGPT • もはやAutoGPTで...

24.

Agenda News https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

25.

Next

26.

Next

27.

Agenda News https://www.ai-tech-c.jp/generative-ai-study-group-gasg/

28.

News • みんなで生成系AIブックマーク • 新企画です!! 皆様の気になった生成系AIの情報をその場で募集して ディスカッションします。 • 募集はこちら https://forms.gle/kKVsb9jDwMTm2Rkk6 から事前投稿も受 け付けています。 • リスト • https://docs.google.com/spreadsheets/d/1bjJfAhfEvNMSuxAszCiSo0qdE89s p86ooAyBdRVuDO0/edit?usp=sharing

29.

EOF https://www.linkedin.com/in/kunihiro-sugiyama-49b0372a/ https://www.ai-tech-c.jp/generative-ai-study-group-gasg/