mlthoughts2018

431 ms·

2941.
by mlthoughts2018 8y ago
Is it reasonable to believe Pigeon could only add value if the time delay between a new delay-causing issue and some verifiable / consensus info getting posted by users is faster than similar updates by an official source like MTA? (I
2942.
by mlthoughts2018 8y ago
I don’t work in the “high-tech HR” space, but do have experience with embedding vector methods, joint-embeddings, and trying to solve generic information retrieval tasks, where the underlying data is not text or images, by using embedding m
2943.
by mlthoughts2018 8y ago
The vector of skills vs. vector of job properties thing could be a coherent idea. I'd argue it's not a good idea (and I have created joint-embedding models for multi-modal text and image nearest neighbor search systems). But it co
2944.
by mlthoughts2018 8y ago
My experience hiring machine learning talent over several years has been that people over-hype the cost of a false positive. Both the article's false positive (expending the cost of interviewing on someone who ultimately reveals to be
2945.
by mlthoughts2018 8y ago
For just this reason I applied for a federal research grant to begin encoding NP complete problems into watchlist format. I can take the Clay Institute money I’m sure to win and blow it all on hats.
2946.
by mlthoughts2018 8y ago
I wonder if there should be a similar law to the “right to repair” automotive laws, but for general software installation or removal for other devices? If a company both controls the production of the device and the sole means by which a us
2947.
by mlthoughts2018 8y ago
This technique generally (applying NLP to 10K’s and other corporate filings) has been in practice for many years. One common, highly commoditized text signal is earnings surprise: measuring sentiment in earnings call transcripts, and compar
2948.
by mlthoughts2018 8y ago
If the quantile assignments took into account data from both the training set and the test set, then it means the training set quantile label itself contains lookahead information about what the distribution of test set values is. This can
2949.
by mlthoughts2018 8y ago
If it’s a serious company, it’s just an expected, standard part of negotiating, no different than salary, bonus, benefits, vacation, etc. In general, you should ask if you’ll be required to sign any NDA, non-compete, or non-disparagement ag
2950.
by mlthoughts2018 8y ago
This is not very convincing. It reads like a semester project for an undergrad or a summer intern. In particular, the prediction task used for evaluation is not very meaningful, and the intended use case (predicting biotech underperformers)
2951.
by mlthoughts2018 8y ago
Do you know if this revenue figure accounts for the “top line revenue problem” (inclusion of government charges, refunds and fare promotions in top line revenue) mentioned for the 2017Q4 numbers at the 13th Naked Capitalism article [0] on U
2952.
by mlthoughts2018 8y ago
I think this might also be supported by looking at some of the better funded / successfully exited health tech startups. I’m thinking of things like Flatiron Health, BenevolentAI, Quartet and Spreemo, and surely many more. At heart the
2953.
by mlthoughts2018 8y ago
> “Scrum requires the TEAM to be cross functional for the product it develops. It does not require each TEAM MEMBER to be cross-functional.” It’s naive to approach this in some letter-of-the-law way. Whatever you want to argue that “Scru
2954.
by mlthoughts2018 8y ago
> “The message was that there are about 7,000 people in Watson Health today and this was a cost-cutting exercise. 90 days’ notice with 30 days’ severance.” If accurate, that severance benefit is just shameful, and simply not good enough
2955.
by mlthoughts2018 8y ago
> "Not sure what this has to do with Scrum or the low complexity of Scrum, which fully can be explained in a small book or a week of training." Earlier you made a disingenuous comparison of the relative bloat of RUP and Agile R
2956.
by mlthoughts2018 8y ago
> "State which specific practices work better than Scrum in specific situations." Sure. - When working on a team whose primary outputs are highly specialized research, the necessity to have planning meetings, standups, retrospe
2957.
by mlthoughts2018 8y ago
Except that Scrum added time-wasting extra stuff on top of the situation. If management doesn't buy in and I can operate my own low-overhead, low-formalism workflow, then so be it. I suffer the badness of management, but I don't
2958.
by mlthoughts2018 8y ago
Generally the problem is that management wants this idea you call "just in time planning" to happen all the time, and even believes it to be synonymous to Agile. Meanwhile the software team is saying, that's fundamentally at
2959.
by mlthoughts2018 8y ago
When it is sought by the team, consensus-wise, "from below" -- I would absolutely support it. That is the team telling you what works for them, not the management telling the team how it must work.
2960.
by mlthoughts2018 8y ago
One example which I actually used to great effect in a previous team was that the team decides the relative allocation of backlog items between new features, fundamental research, maintenance, and refactoring / architectural design.
2961.
by mlthoughts2018 8y ago
I wish more Scrum-advocates shared your opinion. Hey, if Scrum is working for some team, I would be the last person to ask them to change it. I know how much I hate it when I have a workflow that is succeeding and then management arbitraril
2962.
by mlthoughts2018 8y ago
> Being able to chunk complex projects into simpler components that can fit into smaller timeframes... is an essential skill for a software developer. Why do you feel this claim is related to Agile or Scrum in any way? When I see that cl
2963.
by mlthoughts2018 8y ago
> "If you go by actual behavior, it turns out the highest priority at most companies is not actually to improve, to get better at making things for users. It's instead to make managers and executives feel like something is bein
2964.
by mlthoughts2018 8y ago
> "How about deciding what you want to say: a) scrum makes it worse or b) scrum does not help?" Both! > "Really? Before we have worked with RUP and Agile RUP. Compared to that Scrum is tiny." Nuclear submarines are
2965.
by mlthoughts2018 8y ago
I'm comparing Scrum to customized, situation-specific workflow practices created and adopted by the team that will use them, based on what works for that team. What about Scrum uses evidence to convince a team to believe it should devi
2966.
by mlthoughts2018 8y ago
> "It definitely doesn't happen "all the time" in development." It absolutely does in different business units. I'm glad this hasn't happened to you in your business units. That doesn't say anythin
2967.
by mlthoughts2018 8y ago
I agree we might be at the point of trading anecdotes. I would suggest though that there could be the problem that a team works hard to produce good outputs in spite of Scrum, rather than because of it . My feeling, as I have mentioned e
2968.
by mlthoughts2018 8y ago
> What? Scrum fixes all bad management practices? What are you saying here? I was questioning your original statement that I mistakenly thought Scrum should fix bad management practices. I do not think that, and no part of that is a crit
2969.
by mlthoughts2018 8y ago
Quoting from my parent comment, > "And Scrum is misapplied so often that this clearly applies ..." It's not about always being misapplied. Just that it's misapplied so often that in the larger scale decision making,
2970.
by mlthoughts2018 8y ago
What? I don't assume Scrum should or can fix bad management practices. I'm saying adding Scrum to already-bad management practices is like throwing gas on the fire. Scrum doesn't help. Also, if Scrum is only some kind of spec

More ›