Someone

  • Student

Activity Feed

As long as some portion of the fees charged by the fund are allocated to the developer of the algorithm in which capital is placed and investors are flocking to the platform, there will be an incentive to develop algorithms for the platform. However, after reading this article, I considered the perspective of the investor and I was left wondering – what is the advantage to parking capital with Quantopian compared to a broad index fund or another investment vehicle? I couldn’t come up with a viable reason to invest in Quantopian, unless the investor fundamentally believes that one or some combination of algorithms at Quantopian will yield higher returns than the market.

In this case, I don’t think there is a reason to believe that Quantopian will yield higher than average returns. As DIY trading platforms become commonplace, there is less incentive for investors to use the platform and consequently less incentive for developers to produce content for the platform.

On November 14, 2018, Someone commented on Flying High: GE’s Billion Dollar Bet on Additive Manufacturing :

This was a really interesting article – in part because I hadn’t considered the design implications of additive manufacturing. In particular, I think it’s striking that GE was able to effectively eliminate the need for a high number of subtractive parts, while also reducing weight of the component and improving efficiency with additive manufacturing. In my mind, this highlights the value of AM.

I would question the notion that that GE should pursue development of “convenient, multiple-use printing capabilities” to assist with training implications of AM technology for two reasons. 1) I wonder how much printing processes differ for AM when switching between types of printing materials – in which case – is it even feasible to print plastics and metal or alloy based components on the same machine? 2) There are a number of other producers of competing additive manufacturing technologies. Differentiation in the type of machine, software and function will widely vary until there are higher levels of consolidation in this space. Will GE’s investment in this area truly function as a method of standardization or would the company be better served to simply train experts on AM tech based on type of material printed?

Training data fed to machine learning algorithms that make up these models becomes incredibly important in this situation for many of the reasons pointed out by the author above. If the model is trained to assess accuracy on the basis of demographic information and the historical decisions of judges which are inherently exposed to human biases – the model will reflect all of those biases in its assessment. Moreover, it’s not clear to me how this type model can actually react to information that has come to light over the course of the case (where there is often different sets or facts or at minimum different interpretation of facts presented), rather the judgement formed by these types of models seem to be based on historical data and demographics, which to me feels less relevant here when we are assessing facts presented by parties for accuracy and measuring those against ever changing legal standards. For these reasons, I’m skeptical that machine learning would be anything but detrimental in this current application.

I agree with the previous comment. The facilities that stand to gain the most from these CDSS and similar models are those that are likely under-resourced. The subset of training data for models in patient populations is a huge and interesting issue in healthcare IT. Particularly in the case of rare disease, individual health systems may not have broad enough patient populations that have a specific condition in order to accurately train a model to identify the condition and its variants or to recommend specific care pathways.

Another question raised by this article – and by machine learning applications in healthcare more broadly is – how good does a model need to be in order to be ‘good enough’ to trust its results? There’s a general public perception (evident in backlash to relatively rare accidents caused by autonomous vehicles) that we expect algorithms to be perfect when human life is involved. In reality, models should only need to be better than the alternative of MD diagnosis before we adopt them. I personally feel the application of these types of technologies in the next 5-10 years is more likely to be augmented/intelligent decision making on basis of machine learning data interpreted by providers rather than purely prescriptive AI in the absence of MD oversight.

On November 14, 2018, Someone commented on Cedar: Making Paying Doctors Much Easier :

I agree with the suggestion that Cedar should consider incorporating insurance communication to the patient as a part of their platform because that this maintains the core competency of their product on simplify the patient’s understanding of cost of care and easing payment process. One of the primary reasons patients may have difficulty paying and interpreting medical bills today is because of the lack of central billing ‘authority’. In one given patient encounter, coming into contact with a few different providers in a hospital could entail multiple bills as independent physicians may choose to bill for professional fees separately from the facility (hospital fees). This is complicated by the fact that the insurance company may act as an intermediary and cover some portion of these bills, resulting in a plethora of statements that are difficult for the patient to navigate.

The article also left me wondering about Cedar’s own system for driving revenue growth. I think there is a value proposition that Cedar could provide to health systems by suggesting profit sharing arrangements for increases in collections.

On November 14, 2018, Someone commented on Innovation through acquisition at General Mills :

I’m struggling to see how 301 at General Mills can truly serve as a method for open innovation. When I think about open innovation, I think of distribution of responsibility in the typical funnel of product design such that a better design (in this case product design) is achieved by incorporating feedback from variety of outside parties (developers, end-consumers, etc). It’s not clear to me that R&D at General Mills is using the 301 in this fashion. Rather it seems like they are relying on perhaps traditional R&D processes at other firms and are using this VC arm to just acquire additional brands that they feel hold potential.

There is value in maintaining R&D spend at General Mills to push for a platform for open innovation based on active engagement of the consumer , rather than acquisition of innovation processes at other brands. Both of those components will likely be necessary for the firm’s long-term success.

I agree with a few of the points made above that additive manufacturing is still some ways away from being incorporated into mass production – largely because of the cost of production at this point. I see the primary benefit of additive manufacturing as the precise nature of the component or item being produced, but there are still several limits on the 1) types of materials that can be used in assembly with these machines, 2) the speed at which components can be fashioned, and most importantly 3) the cost of production.

In sports apparel, I feel that the ultimate implication of additive manufacturing (implied by the “Flyknit” portion of the article above where Nike is developing components of a shoe suited to an athlete) is the development of truly custom apparel that’s better suited to individuals physique and style of play.