投稿時間:2021-09-28 04:19:49 RSSフィード2021-09-28 04:00 分まとめ(25件)

カテゴリー等	サイト名等	記事タイトル・トレンドワード等	リンクURL	頻出ワード・要約等/検索ボリューム	登録日
AWS	AWS Database Blog	Implement active/active replication between Amazon Aurora clusters using Oracle GoldenGate	https://aws.amazon.com/blogs/database/implement-active-active-replication-between-amazon-aurora-clusters-using-oracle-goldengate/	Implement active active replication between Amazon Aurora clusters using Oracle GoldenGateEnterprises both large and small across diverse industries and with varying levels of cloud maturity recognize the importance and value of deploying active active database configurations An active active system is a network of independent processing nodes each having access to a common replicated database so all nodes can participate in a common application Some enterprises are …	2021-09-27 18:20:13
海外TECH	Ars Technica	2021 iPad mini suffers from uneven “jelly scrolling” in portrait mode	https://arstechnica.com/?p=1798690	hardware	2021-09-27 18:25:28
海外TECH	DEV Community	What's your favorite JS testing framework?	https://dev.to/madza/what-s-your-favorite-js-testing-framework-5c2l	What x s your favorite JS testing framework Nowadays the industry is dominated by agile development and tight deadlines Manual unit testing can be complex and time consuming so it makes sense to rely on automated frameworks Some of the favorites include Jest Mocha Jasmine Karma Do you unit test your code and what s your favorite testing framework	2021-09-27 18:52:47
海外TECH	DEV Community	Not Feeling Like Enough When in Tech	https://dev.to/deepeshaburse/not-feeling-like-enough-when-in-tech-234c	Not Feeling Like Enough When in TechMost of us if not all feel as though we re not enough at some point in some context Since I ve decided to make a career in tech I have felt so at multiple points and so thought of writing about it This blog is my point of view how I deal with the feeling and what I would expect from the other person if I were to express to someone My sole aim of writing this blog is to start a conversation and not just be more in touch with our feelings but also become more sensitive towards others Social media has definitely made a huge impact on me and the large number of amazing supportive tech communities is something that I am so grateful for Unfortunately just like everything else I feel like the presence of so many communities and social media platforms is a huge contributor to this feeling I ll feel perfectly fine until I see this one tweet saying someone has learnt some new technology or a post on LinkedIn talking about this exceptional internship they got a famous company and although I am very happy to read such a news it starts the Am I doing enough Can I achieve that loop It makes me criticize everything I ve achieved yet and if I really belong to this field It s not a good feeling to say the least Even though I absolutely understand why we feel this way and it is very tough to think of it in any other way it is really worth considering what we re comparing to and who Tech is such a vast field New things are being made launched every day we are making progress in leaps and bounds and we need to be mindful of that Keeping up with that and expecting ourselves to know everything is honestly a little unfair to ourselves As long as we re making some sort of progress every day and that includes breaks every once in a while especially when we need one scrutinizing yourself over every little thing is not a good idea Another little thing I love to remind myself is that someone will always be better than me and I will always be better than someone And this is also very contextual I might be better than someone in one technology and that same person may be so much better than me in some other technology Talking to loved ones or anyone you are close to helps a lot too They remind you of the efforts you are putting in and that at the end of the day that s all that matters I have been very lucky with the people in my life they have always been very supportive of me and not just reminded me that I m good at what I do but also that I am much more than just my career It s a great reminder that our career is very important but it is not the only thing that matters Even though we feel this I think we are not as sensitive to others or rather considerate that others go through this too When someone confides in you that they feel this way the most common reply I ve seen is Don t worry you ll get there some day and although that s a nice thing to say I don t really think that helps Talking from experience instead of being told I will probably get there some day I would love one addition to that sentence You re doing great right now don t compare yourself to them but if you do wish to achieve what they did I m sure you can and will get there some day When someone is already feeling low one you re doing great can go a long way especially when they trust you with their feelings And in case no one has told you today I think you re doing great I hope this article pushes you to be kinder to yourself and those around you I would love to hear how you all cope with this feeling please drop a comment and share how you do it might help someone	2021-09-27 18:23:15
Apple	AppleInsider - Frontpage News	Apple engineers talk about iPhone 13 camera changes	https://appleinsider.com/articles/21/09/27/apple-engineers-talk-about-iphone-13-camera-changes?utm_medium=rss	Apple engineers talk about iPhone camera changesAn interview with a number of Apple vice presidents involved with the engineering of the iPhone camera system has been published proving more insight into the decisions behind the improvements for the releases Published on Monday the iPhone Talking to the Camera Engineers episode of the Stalman Podcast features a trio of Apple representatives The group is headed up by Kaiann Drance VP of Worldwide iPhone Product Marketing along with VP of Camera Software Engineering Jon McCormack and VP of Camera Hardware Engineering Graham Townsend For the iPhone Apple has brought its Sensor Shift OIS as well as improvements to low light photography Photographic Styles and Cinematic Mode On the Pro models there s a new Macro mode along with support for ProRes video Read more	2021-09-27 18:58:11
Apple	AppleInsider - Frontpage News	iFixit iPhone 13 Pro teardown reveals Face ID changes, display swap issues	https://appleinsider.com/articles/21/09/27/ifixit-iphone-13-pro-teardown-reveals-face-id-changes-display-swap-issues?utm_medium=rss	iFixit iPhone Pro teardown reveals Face ID changes display swap issuesRepair site iFixit has completed its full teardown of the iPhone Pro revealing additional details about the TrueDepth sensor and the ability for third parties to swap parts Credit AppleThe repair specialists at iFixit first completed an initial teardown of the device on Sept Now iFixit has published part two of the teardown focusing on other changes and the overall repairability of the devices Read more	2021-09-27 18:39:47
Apple	AppleInsider - Frontpage News	Apple TV+ promotes 'Wolfboy and the Everything Factory' with educational trailers	https://appleinsider.com/articles/21/09/27/apple-tv-promotes-wolfboy-and-the-everything-factory-with-educational-trailers?utm_medium=rss	Apple TV promotes x Wolfboy and the Everything Factory x with educational trailersApple has released a trailer for Wolfboy and the Everything Factory the Joseph Gordon Levitt animated series that started streaming on Apple TV on September The trailer posted to the Apple TV YouTube account on Monday is titled Intro to Adventures with Professor Luxcraft The clip has the titular professor teaching the viewer about Sprytes creatures within the show s forest that can create things for the human realm It appears that the YouTube channel has mistakenly uploaded the Sprytes video with the description and subtitles for the similar Adventure clip while the reverse occurred on Friday for that day s video Read more	2021-09-27 18:20:05
海外TECH	Engadget	Democratic lawmakers say Facebook 'must completely abandon' Instagram Kids	https://www.engadget.com/democratic-lawmakers-cancel-instagram-kids-184621575.html?src=rss	Democratic lawmakers say Facebook x must completely abandon x Instagram KidsMere hours after Facebook said it was pausing work on Instagram Kids a group of Democratic lawmakers has called on the company to instead completely abandon the project Senators Ed Markey and Richard Blumenthal and Representatives Kathy Castor and Lori Trahan say Facebook s decision to halt development is “insufficient Facebook is heeding our calls to stop plowing ahead with plans to launch a version of Instagram for kids But a quot pause quot is insufficient Facebook must completely abandon this project ーEd Markey SenMarkey September “Facebook has completely forfeited the benefit of the doubt when it comes to protecting young people online and it must completely abandon this project the group said in a joint statement on Monday They re the same four lawmakers who told Facebook earlier in the year they had “serious concerns about the project when they first learned about it Facebook said it was suspending work on Instagram Kids after The Wall Street Journalpublished a report that claimed the company had ignored its own research on the harm apps like Instagram can do to young people The company quickly refuted that piece saying its studies showed young people could have both positive and negative experiences interacting with social media The company will have to answer questions on its research later this week when the Senate Commerce Science and Transportation Committee holds a hearing on the matter on Thursday	2021-09-27 18:46:21
Cisco	Cisco Blog	Cisco Secure Firewall: Sometimes, updates are great news	https://blogs.cisco.com/security/cisco-secure-firewall-sometimes-updates-are-great-news	gains	2021-09-27 18:52:17
Cisco	Cisco Blog	Using INFRAM to Modernize Today’s Healthcare	https://blogs.cisco.com/customerexperience/using%e2%80%afinfram%e2%80%afto-modernize-todays-healthcare	Using INFRAM to Modernize Today s HealthcareInvestments in infrastructure are on the rise fueled by the digital transformation of healthcare A foundational digital infrastructure is key to the modern healthcare organization of today and tomorrow But many healthcare leaders are underutilizing the technology they have or may not have visibility where there are gaps in their strategy Learn how INFRAM can help	2021-09-27 18:48:25
海外科学	NYT > Science	Sherwood Boehlert, a G.O.P Moderate in the House, Dies at 84	https://www.nytimes.com/2021/09/24/us/sherwood-boehlert-dead.html	Sherwood Boehlert a G O P Moderate in the House Dies at A champion of environmentalism who chided climate change skeptics he was among the last of the relatively progressive Rockefeller Republicans	2021-09-27 18:46:06
ニュース	BBC News - Home	Fuel demand should return to normal in coming days, say suppliers	https://www.bbc.co.uk/news/uk-58707455?at_medium=RSS&at_campaign=KARANGA	sources	2021-09-27 18:43:57
ニュース	BBC News - Home	Labour conference: Frontbencher Andy McDonald quits in protest at Sir Keir Starmer	https://www.bbc.co.uk/news/uk-politics-58713175?at_medium=RSS&at_campaign=KARANGA	starmer	2021-09-27 18:54:14
ビジネス	ダイヤモンド・オンライン - 新着記事	ファーウェイ幹部解放に見る米中の実利的対応 - WSJ PickUp	https://diamond.jp/articles/-/283140	wsjpickup	2021-09-28 03:50:00
ビジネス	ダイヤモンド・オンライン - 新着記事	シリコンバレーの若い億万長者がもたらす社会への還元 - シリコンバレーの流儀	https://diamond.jp/articles/-/283052	億万長者	2021-09-28 03:45:00
ビジネス	ダイヤモンド・オンライン - 新着記事	北京冬季五輪のコロナ対策、東京より厳格に - WSJ PickUp	https://diamond.jp/articles/-/283141	wsjpickup	2021-09-28 03:40:00
ビジネス	ダイヤモンド・オンライン - 新着記事	コロナ禍の中南米脱出、米国への移民が多国籍化 - WSJ PickUp	https://diamond.jp/articles/-/283142	wsjpickup	2021-09-28 03:35:00
ビジネス	ダイヤモンド・オンライン - 新着記事	定年後に転職するために必要な、2つのこと - 自分だけは損したくない人のための投資心理学	https://diamond.jp/articles/-/283139	魅力	2021-09-28 03:30:00
ビジネス	ダイヤモンド・オンライン - 新着記事	「つい良い人を演じてしまう」自分にイライラ、解決法は？ - イライラ・モヤモヤ職場の改善法　榎本博明	https://diamond.jp/articles/-/283138	差し障り	2021-09-28 03:25:00
ビジネス	ダイヤモンド・オンライン - 新着記事	病室に患者用無料Wi-Fi導入がやはり必要な理由 - News&Analysis	https://diamond.jp/articles/-/283184	病室に患者用無料WiFi導入がやはり必要な理由NewsampampAnalysis月、全国の地域の拠点となる病院の病室で「患者が無料WiFiを使えるかどうか」民間団体が調査し、都道府県別の導入率ランキングと病院名を公表した。	2021-09-28 03:22:00
ビジネス	ダイヤモンド・オンライン - 新着記事	首都圏「中高一貫校」最新人気情勢【2022年・男子受験生版】 - 中学受験への道	https://diamond.jp/articles/-/282955	中学受験	2021-09-28 03:20:00
ビジネス	ダイヤモンド・オンライン - 新着記事	ひろゆきが「この人なら一緒に働きたい！と思う人」ベスト1 - 1％の努力	https://diamond.jp/articles/-/282658	youtube	2021-09-28 03:15:00
ビジネス	ダイヤモンド・オンライン - 新着記事	文章の最終確認は「パソコン」ですべき？「紙」ですべき？ - 短いは正義	https://diamond.jp/articles/-/282753	資料	2021-09-28 03:10:00
ビジネス	ダイヤモンド・オンライン - 新着記事	「新しい分野を独学するとき」の原理原則を学べる意外な一冊 - 独学大全	https://diamond.jp/articles/-/281424	読書	2021-09-28 03:05:00
GCP	Cloud Blog	Google Cloud Vertex AI + Battlesnake: Using practical reinforcement learning to defeat your friends	https://cloud.google.com/blog/topics/developers-practitioners/google-cloud-vertex-ai-battlesnake-using-practical-reinforcement-learning-defeat-your-friends/	Google Cloud Vertex AI Battlesnake Using practical reinforcement learning to defeat your friendsHow do you like to learn a new skill Read books Take a course Try to build something We are a small group of Customer Engineers who with no formal background in machine learning who decided to learn about reinforcement learning by entering Battlesnake an online survival game where our trained agent would cut its teeth against competitors around the world  Our project was fueled by Vertex AI Google Cloud s machine learning platform and we share our process and lessons learned here in the hope it may inspire you to also learn by doing Let s consider a different approach utilizing the construct of a game to evaluate new technology and learn new skills   Enter the arenaBattlesnake isn t your indestructible Nokia candy bar CDMA phone snake game This isn t even an updated Google Snake spin off but do try and get the secret rainbow snake this is something very different and much more useful On the surface Battlesnake seems like a simple game with a small number of basic rules Don t run into walls or other snakesDon t starveDon t get eaten by another snakeOnce you break through the basic premise you ll soon realize it is a lot more complicated than that There are many ways to build and place your own battlesnake into a competition Depending on your team s experience level you may want to try out one of the starter projects that Battlesnake makes available  Alternatively you may want to start wading into the deeper end of the competitive pool and enhance your snake with health based heuristics models or cannonball into the pool using a reinforcement learning approach The approach we took to our first competition was to hedge our bets a little get something into competition quickly and gather some data to iterate on then explore improvements on the initial snake performance through a series of ML model tweaks ultimately building a reinforcement learning model that we were sure was going to win in the most virtuous and collaborative sporting way of course  More on results later but here is walkthrough of how our architecture and development progressed Introduction to reinforcement learningReinforcement learning often referred to as RL has had a long history as a way to build AI models From games like chess Go and Starcraft II to more industry specific problems like manufacturing and supply chain optimization reinforcement learning is being used to build best in class AI to tackle increasingly difficult challenges  For those unfamiliar with RL here is a quick primer Traditionally machine learning models learn to make predictions based on massive amounts of labeled example data In RL agents learn through experimentation Each iteration is scored based on a reward function As an example for Battlesnake a basic set of rewards might be a for winning and a for losing  The rewards are fed into the model so that it “learns which moves earn the highest reward in any given scenario Similar to humans learning to not touch a hot stove the model learns that running a snake head first into a wall will produce a negative reward and the model will remember not to do that most of the time  For complex systems this reward structure might consist of dozens of different inputs that help to shape the reward based on the current state of the overall system Our team did not have a classically trained machine learning expert but we did have enough expertise to take some concepts that we learned from others who had attempted this approach and apply them using Google Cloud s Vertex AI platform How wecharmedtrained our snakeOne of the key starting areas for building a RL model is to set up an environment that knows how to play the game OpenAI s gym toolkit provides an easy way for developers to get started building RL models with a simple interface and many examples to start training your model quickly This allows you to focus purely on the parts of the model that matter like For our Battlesnake competition we had one week to build train and test a snake before throwing it into the arena against our peers in a winner takes all tournament Though traditional ML loops can take weeks or months to build with the power and simplicity of Vertex AI we made an end to end ML pipeline in a few hours This velocity freed up several days to run training experiments and tune our model  Training on Vertex AI in this case started with a custom training job Vertex AI allows you to specify your own training container in our case we used Pytorch to run any arbitrary training tasks on the Vertex AI managed infrastructure Initial experimentation started in a basic Jupyter notebook hosted on Vertex AI but we quickly transitioned to a custom Docker container that was produced by Cloud Build on a push to our main source repository We attempted to source our container image from multiple prebuilt images but ultimately found that we were spending too much time working through conflicting dependencies which was slowing down our progress so we switched to a cleaner base image containing just the NVIDIA CUDA drivers As we started to scale our training we converted our notebook into an ML pipeline using Vertex AI s managed pipeline service The workflow took advantage of the built in hyperparameter tuning service to automatically tune our model to perform as well as possible with minimal supervision from our team Each training job ran for hours before going into the evaluation phase During evaluation we selected previous versions of the model to run simulated games against to see if the newly trained model performed better than previous versions The winner of the evaluation battles was promoted to the top model and the new baseline for the next iteration of training Our pipeline was triggered on demand to do additional training This was usually after we loaded the new model into the snake in the global arena and observed its behavior If we were taking this fully to production we would have used the data we captured about game win rates to intelligently trigger additional training if the model was starting to underperform Deployment architectureWe deployed our top model to production using Vertex AI s model serving infrastructure The data sent from Battlesnake is a JSON document of all the current positions of the snakes the food and any hazards that are on the board Responses from our server were required to be completed in ms Any snake that “timed out responding to requests from the game would continue moving in the same direction that it was already traveling Given the time constraint we took a two pronged approach to control our snake Requests from Battlesnake first came to a web service deployed on Google App Engine The App Engine server forwarded the request to the model server with a hard timeout of ms Responses from the model were validated in the App Engine server as a last fail safe in case the model responded with a move that was clearly invalid move outside of the board etc If the model service timed out the App Engine server fell back on its own logic to produce a reasonable move Data analysisOur snake played thousands of games within the Battlesnake ecosystem both in the public arena and our private tournament Each game consisted of tens to hundreds of moves and produced a finite result As part of our experimentation we captured all the messages sent from the Battlesnake platform and fed them via Pub Sub to BigQuery to be used later for statistical analysis of our performance As mentioned in the previous sections the data received from Pub Sub was not in the standard tabular format but in an array of JSON objects To perform analysis we first needed to convert the data into a standard table format We leveraged the UNNEST function in BigQuery to flatten the record into the table The data points in the messages from the battles consisted of the health length food location head and body coordinates turns and latency for all the snakes on the board including board size etc For exploratory analysis of our ML model we attempted to answer questions like which snake did we battle the most and how did we perform in each of those battles Is there a particular snake that keeps beating us What are the frequent food locations After gathering these insights we can further refine how we tune our model parameters to what works best for our snake ConclusionOur snake performed very well in open competition in the public arena but had a string of bad luck in actual tournament play During the group stages we competed against two of the snakes that made it into the finals The primary snake in our group was a hungry monster that quickly gobbled up food and boxed our snake out very effectively This led to our untimely death by wall collision and a fast exit from tournament play We accomplished a reasonable amount of success in the global arena for a very short amount of training time thanks to the power of Vertex AI  To learn moreBattlesnakeVertex AI ML Training Managed Pipelines Hyperparameter Tuning Model ServingMachine Learning	2021-09-27 18:30:00