[{"data":1,"prerenderedAt":251},["ShallowReactive",2],{"project-21962326":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":35,"aspects":36,"strip":41,"post_text":43,"what":44,"said":71,"maker_in_thread":158,"shape":166,"grid":170,"attempts":214,"same_week":215,"maker_shows":242,"maker_hidden":243,"reader_version":244,"card":245},"21962326","bangBangML","Self-taught neural network learns to shoot a target","title","Show HN: Self-taught neural network learns to shoot a target","https:\u002F\u002Fgithub.com\u002Fvictorqribeiro\u002FbangBangML","github.com","atum47",1578244513,2020,60,21,980,3028,33,"notable","ai","neural network demo","web_app",[24,25,26,27],"neural network","javascript","reinforcement learning","game ai","debated",-0.227,{"positive":31,"neutral":32,"mixed":33,"critical":34},4,2,0,5,11,{"idea":37,"execution":39,"performance":40},{"positive":38,"critical":31},1,{"positive":38,"critical":33},{"positive":33,"critical":38},[38,34,33,33,33,33,33,34,33,34,34,33,42,38,33,33,38,38,42,33,38],3,null,{"problem":45,"audience":43,"pricing":43,"motivation":56,"time_to_build":43,"stack":62,"openness":70,"stage":70,"team_size":70},{"value":46,"source":47,"evidence":48},"Demonstrates a neural network learning to aim and hit a target","inferred",[49,53],{"id":50,"by":51,"text":52},"21962953","ssivark","Not unlike most neural network solutions :-P But hey, this seems to be a fun project more than anything else. And working in such low-dimensional spaces with an obvious…",{"id":54,"by":11,"text":55},"21963794","think about it like this: with a 10 minutes training the neural network takes about 8 shots to hit the target, so as the player, you have 8 shots to hit it before it hits you.…",{"value":57,"source":58,"evidence":59},"Fun project; idea for game difficulty levels using a trained network","maker",[60,61],{"id":54,"by":11,"text":55},{"id":50,"by":51,"text":52},[63],{"value":64,"source":47,"evidence":65},"JavaScript",[66],{"id":67,"by":68,"text":69},"21965220","reubens","I'm curious about the title of the module 'Dejavu.js'. Is it from another project?","unknown",{"praise":72,"objections":89,"requests":119,"questions":137,"alternatives":149},[73,78],{"value":74,"source":75,"evidence":76},"Fun project and useful for beginners to build intuition","commenter",[77],{"id":50,"by":51,"text":52},{"value":79,"source":75,"evidence":80},"Encouragement to finish and ship the idea",[81,85],{"id":82,"by":83,"text":84},"21964808","rs23296008n1","Never just an idea. Remove the \"just\". It is a realised idea and it was important that you execute the idea and, to whatever useful extent, finish it. Don't allow that negative…",{"id":86,"by":87,"text":88},"21966214","stanfordkid","On point! So much harder to execute and create something real and tangible than pontificate.",[90,97,108],{"value":91,"source":75,"evidence":92},"Overengineered for the problem",[93],{"id":94,"by":95,"text":96},"21962926","eerrt","Am I the only one who thinks this is a little overengineered?",{"value":98,"source":75,"evidence":99},"Not really self taught since the environment provides feedback and the rule is hard-coded",[100,104],{"id":101,"by":102,"text":103},"21970917","Nychtelios","The environment is the training data and you are forcing the NN to hit the target. It isn't self taught. No.",{"id":105,"by":106,"text":107},"21963005","bluesign","It doesn't look like totally self taught though \"If the shot was to the left of the target, adjust the weights to the right and vice and versa. On this step I'm not saving the…",{"value":109,"source":75,"evidence":110},"Equivalent to a simple PID controller or polynomial fit",[111,115],{"id":112,"by":113,"text":114},"21963885","p1necone","If it's done like that isn't it just an overly complex, not very good PID controller (with one or more of those letters missing)?",{"id":116,"by":117,"text":118},"21964598","mam2","So you learnt a polynomial of order 2. Great job, I'm quite scared of the killer robots now.",[120,126],{"value":121,"source":58,"evidence":122},"Add wind strength and direction as in Worms",[123],{"id":124,"by":11,"text":125},"21963054","yeah, I'm thinking about stealing wind strength and direction from the Worms game",{"value":127,"source":75,"evidence":128},"Reference wind physics from Gorillas and Gunbound",[129,133],{"id":130,"by":131,"text":132},"21963851","verst","Remember Gorillas from the QBasic language? That had wind speed and direction. Perhaps helpful. https:\u002F\u002Fgithub.com\u002FGorillaStack\u002Fgorillas\u002Fblob\u002Fmaster\u002Fgorillas.bas",{"id":134,"by":135,"text":136},"21968099","goldenkey","Gunbound was extremely complex with all kinds of variables including boomerang style projectiles.",[138,145],{"value":139,"source":75,"evidence":140},"Will the network learn to aim at target edges?",[141],{"id":142,"by":143,"text":144},"21963257","thomasfedb","I wonder if it will teach itself to aim at the edges of the target zone, given that's where the \"adjust to the right and vice versa\" method will end up landing shots.",{"value":146,"source":75,"evidence":147},"Is the Dejavu.js module from another project?",[148],{"id":67,"by":68,"text":69},[150,154],{"value":151,"source":75,"evidence":152},"Gorillas (QBasic game with wind)",[153],{"id":130,"by":131,"text":132},{"value":155,"source":75,"evidence":156},"Gunbound",[157],{"id":134,"by":135,"text":136},{"comments":159,"answered":42,"key":160},6,[161,162,165],{"id":124,"text":125},{"id":163,"text":164},"21963036","probably not",{"id":54,"text":55},{"commenters":167,"max_depth":31,"maker_replies":159,"first_comment_mins":168,"lifespan_hours":169},16,74,49,[171,173,175,177,179,180,181,182,184,187,189,192,196,199,200,203,205,206,208,210,213],[94,33,38,172,95,33],"project\u002Fcritical\u002Fidea",[50,33,34,174,51,33],"project\u002Fpositive\u002Fidea",[124,33,33,176,11,38],"project\u002Fneutral\u002Fidea",[130,33,33,178,131,33],"other\u002Fneutral\u002Fnone",[134,33,33,178,135,33],[163,33,33,172,11,38],[54,33,33,176,11,38],[82,33,34,183,83,33],"maker\u002Fpositive\u002Fnone",[185,33,33,186,11,38],"21965501","maker\u002Fneutral\u002Fnone",[86,33,34,188,87,33],"maker\u002Fpositive\u002Fexecution",[190,33,34,183,191,33],"21969045","jeffshek",[193,33,33,194,195,33],"21982789","commenter\u002Fneutral\u002Fnone","armatav",[197,33,42,176,198,33],"21970717","darepublic",[105,38,38,172,106,33],[201,38,33,202,11,38],"21963033","maker\u002Fneutral\u002Fidea",[101,38,33,204,102,33],"commenter\u002Fcritical\u002Fidea",[112,38,38,172,113,33],[142,32,38,207,143,33],"project\u002Fcritical\u002Fperformance",[67,42,42,209,68,33],"project\u002Fneutral\u002Fnone",[211,42,33,212,11,38],"21965258","project\u002Fcritical\u002Fexecution",[116,31,38,172,117,33],[],[216,221,225,229,234,238],{"id":217,"name":218,"points":219,"category":220},"21991382","Beef",533,"languages",{"id":222,"name":223,"points":224,"category":20},"21934247","Confusing Coleopterists",308,{"id":226,"name":227,"points":228,"category":220},"21951692","amazon-kinesis-video-streams-webrtc-sdk-c",251,{"id":230,"name":231,"points":232,"category":233},"21981158","jql",217,"dev_tools",{"id":235,"name":236,"points":237,"category":233},"21958986","Terrastruct",187,{"id":239,"name":240,"points":241,"category":220},"21948472","sscaffold-css",158,53,false,"v2",{"id":246,"name":5,"line":6,"ai":243,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":247,"top":18,"kw":248,"maker":11,"domain":10,"tier":19,"attempts":33,"strip":249,"url":9,"mix":250,"praise":74,"objection":91,"sub":21},21962326,-0.23,[24,25,26,27],"150000050550310011301",[31,32,33,34],1794036585129]