Interviews

GGPredict is trying to revolutionise CS:GO teaching with AI

Whether or not you’re an newbie Counter-Strike: International Offensive participant trying to climb the rankings, or trying to enhance particular person and crew efficiency in an expert setting, coming throughout digestible, deep-level statistical evaluation could be difficult. 

Teaching and VOD evaluation, whereas helpful, can usually be a time-consuming affair that’s extra scenario-based and qualitative, or primarily based on very top-level information that’s available. 

Pushed out of a want for self-improvement by means of measurable information – GGPredict is an AI-driven software that makes use of quantitative evaluation and machine studying to present customers exact, simply viewable suggestions on their CS:GO abilities. 

Esports Insider sat down with Co-founder Przemysław Siemaszko to debate GGPredict’s journey to the launch of its reside beta on September 24th, what folks have liked about it up to now, and what’s subsequent on the playing cards for the Warsaw-based startup. 

Credit score: GGPredict

RELATED: Digital Creativity Labs develops real-time prediction mannequin for esports

Esports Insider: Would you be capable of inform us a bit about your self and the way GGPredict received off the bottom?

Przemysław Siemaszko: I’m an information scientist and along with a bunch of programmers and other people specialising in information in my earlier firm, we have been enjoying Counter-Strike lots. Ultimately, we realised that given how a lot we play, we’re actually, actually unhealthy and we have to do one thing to enhance. 

Being information folks, our first thought was we have to get statistically sound details about what makes us so unhealthy. We had some concepts, however there have been loads of arguments about who was the worst and who ought to enhance the quickest. We began on the lookout for options that will make us higher and inform us what’s unsuitable with our recreation.

ESI: So you then’ve received the stats to show it.

Pictured: Przemysław Siemaszko. Picture credit score: transient.pl

PS: And to be completely frank we needed to show to considered one of our mates that dashing with P90 will not be the optimum resolution in each spherical. Once we observed there’s no such software obtainable on-line that will shortly inform us what’s unsuitable or what’s proper, we began taking a look at recreation recordsdata and the way the sport processes every thing that occurs. 

Fairly quickly we realised its doable to take information from the sport recordsdata and begin engaged on it. With a few weeks of labor, we realised that mainly every thing within the recreation is in our fingers. We are able to take demo recordsdata or reside video games after which current them in information and numeric format to quantify every thing there may be within the recreation. 

Once we began getting the information out of recreation recordings loads of the stuff we’re accumulating getting will not be doable to see with the bare eye. Clearly, if one participant stood in spawn and one other was operating round their distance differed. However aside from that, you can not measure response occasions, distance lined, accuracy, very tiny variations. That’s the way it received began. 

ESI: We see loads of digital teaching provided on a one-to-one foundation, normally from a professional participant or somebody thought-about an ‘skilled’. How does GGPredict’s AI-driven method set you aside from these rivals?

PS: Our teaching thought got here from the truth that we will predict the outcomes of the sport and see the small print within the information that no person else can. In comparison with a standard coach, now we have far more details about something that occurs within the recreation. 

With a standard coach or professional participant that teaches you, they’d need to rewatch your recreation with you, which already takes an hour they usually have solely a restricted capability together with their different commitments. For us, it takes a few seconds to have a holistic view of every thing that occurred within the recreation. We are able to analyse far more video games and deeper insights. 

Once you watch some video tutorials they advise you “this flash is nice, throw it right here at the moment.” We’re in a position to provide you with particular, situational recommendation on how one can enhance, what about your recreation makes you a great participant and the place you’ll be able to enhance.

ESI: So the suggestions is extra geared in direction of a specific crew arrange or a selected spherical, and throwing the flash on this spot on this context can be the higher choice?

PS: Yeah so on one hand concerning flashes, once we do our evaluation of utilities typically one factor we’d inform you is “okay your flashes are suboptimal, right here’s the way you throw them” and there can be a map with every thing you threw. Then it’ll be in comparison with the positioning of your opponents, in comparison with the a whole bunch of 1000’s of matches we’ve analysed to know the place folks typically place themselves to allow them to clearly see.

That’s only one instance, however now we have extraordinarily personalised data. When giving generic recommendation you understand, “it’s good to defend this passage or this space.” However whenever you play you’ll be able to both stand in apparent areas or you’ll be able to play off sure angles to shock your opponents. 

We are able to measure what the optimum thought is, and if you happen to place your self within the unsuitable positions and you find yourself shedding we will take a look at the place you might be, your issues with positioning and so forth. We are able to counsel coaching maps, take a look at warmth maps after which be taught from that. 

Picture credit score: GGPredict

ESI: Do you get the extent of element right down to peeking or jiggle peeking, is it that granular? Or is {that a} bit more durable to nail down?

PS: We are able to have this information however we’ve determined in opposition to it for now as we really feel it will be a tad overwhelming for gamers. In case your downside is the precise second of peeking then your positioning, tactical responsiveness isn’t actually the problem for you. 

You’d be higher supply taking a look at data concerning your utilities and the way you work together along with your teammates within the recreation. We determined going too deep into element wouldn’t present a lot worth for customers, but it surely’s one thing we’re measuring as properly. 

RELATED: How Athlane intends to simplify esports sponsorships

ESI: How has your partnership with TDJ Pitango allowed you to upscale by means of to your beta launch this month?

PS: They’re our investor and part of Pitango Enterprise Capital which is without doubt one of the largest Israeli funds. Being technical folks, CS:GO gamers, programmers and so forth we didn’t have loads of enterprise know-how. Having somebody like them behind us and giving us enterprise recommendation, doing workshops with us on the best way to run a startup has been actually worthwhile. 

With out them, we wouldn’t have been in a position to get to the place we’re proper now. At this stage, we’re already on the lookout for one other spherical of funding and hopefully, there shall be much more companions like them becoming a member of up with us. 

Credit score: GGPredict

ESI: I do know it’s solely been a couple of days, however have you ever had the uptake you anticipated for the reason that beta launch?

PS: It’s been good up to now, we’re fairly stunned by how many individuals visited the web site and what number of registered, it’s going very properly. Clearly, there have been loads of bugs and programming hiccups that we found out instantly after folks poured into the web site. However that’s the concept behind beta testing, you discover these instantly. 

The beta check that we’re doing proper now’s fairly early stage, we’ve solely launched a few options. We wish to examine if every thing is working properly and we will function many individuals as we would like. 

Now each few days there’ll be new elements and functionalities of the web site and platform being turned on. Hopefully, those who have already registered will proceed having fun with what we’re bringing in and extra folks shall be joyful to affix and get worth out of our product. 

Has there been any commonality in suggestions up to now, some options individuals are responding to properly?

PS: I’d say essentially the most optimistic suggestions we’ve seen is our ‘similarities view’, which is definitely not the primary characteristic of what we’re doing, however provided that now we have so many metrics describing professional gamers playstyle, we did this small algorithm that compares your play to professionals. Then you definitely’ll seem on a desk in comparison with the highest 10 gamers most much like you. 

Individuals are having loads of enjoyable with that, even on the workplace. After each matchmaking or FACEIT recreation we’re enjoying we’re checking, “I’m much like Kirby proper now.” In order that’s one thing individuals are actually having fun with and extra typically, displaying your sturdy and weak sides in a visible type is one thing individuals are having fun with, even now with solely a few of our functionalities switched on.

Picture credit score: GGPredict

After seeing the way you undergo the launch of this reside beta, what are the following large steps for GGPredict? Is it simply rolling out extra options as you go, is it increasing to Dota 2 as has been talked about up to now, or is it one thing else?

Proper now we’re going to concentrate on CS:GO, now we have so many options both prepared or about to be prepared and there’s a lot to do and we’ve received so many concepts that we wish to focus all our efforts on offering the most effective answer for CS:GO. 

The second we really feel that we will supply the most effective you will get in Counter-Strike we’ll transfer onto our subsequent recreation which can more than likely be Dota 2, however we don’t wish to field ourselves in but. 

And is there something particularly that you just’re on the lookout for in the case of new funding companions? 

We have already got some provides and really superior talks with a number of buyers, however we’re not dashing it.  So we’re slowly going by means of the mid-meetings, speaking with buyers from Poland however largely from abroad. Pitango is a superb associate nonetheless and we’d wish to have somebody that not solely brings cash and funding however information and contacts and so forth.

Disclaimer: It is a sponsored piece by GGPredict

Tags

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button
Close
Close