I've been messing with contrastive learning in the CLIP-like "minibatch" style for applications where correspondences between modalities within an example are a relevant source of information, and have been pleasantly surprised by how much this can actually in some cases speed up or otherwise aid with learning different objectives framed as decoding each modality's embedding for a different downstream objective. It seems that tying the representations together in this way can serve as a kind of semi-orthogonal correcting force on the embedding's ability to retain its original reconstructability while not constraining it too much to shift the representation for other goals, and many properties of the resulting models (like the obvious one of having interchangeable downstream decoding functionalities regardless of which modality is used to encode) suggest to me that it might be another tool to use when pursuing a better theory of objective balancing, as well as significant practical use in applications obviously. I'm surprised people aren't making more random contrastive embeddings for sets of "non-obvious" coupled representations
I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week. Does that even exist anywhere without like a PhD from rich people schools? Aside from the moral issues this myopic mindset of businesses just refusing to trust or support their people in any sense sure makes it hard to believe they're actually capable of "innovating" anything. Necessity may be the mother of invention, but focusing on speedrunning exploiting will always leave exploring neglected
> I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week.
Here's my tip: Get a regular mid-level web development job; spend 40% of the week to do the minimum of actual work you were allocated (just try not to dip below that), and spend the rest of your time doing whatever other work you want; tell your boss that while you're finding your allocated work difficult, this additional work that you're doing "out of hours" is really motivating you. The best case scenario is that they'll find that "extra" sufficiently useful to actually change your job title; worst case, you're let go, and can try again in a different company, having gained experience.
I mean I've worked for multiple companies wherein I've been explicitly hired as R&D staff and then let go because they don't want that anymore and would rather downsize than pay another dev to do the short-term work they are now focusing on, and also for companies where the workload wasn't too heavy and left me time to do other stuff but they ran out of runway without telling anyone. I've also worked for a company where I did three people's jobs that weren't mine for three months and got commended for saving the company (not my words) but then unceremoniously laid off because promises were made during that time (that I didn't ask for) and then spreadsheets were consulted and said promises started looking expensive
During none of these instances was I given any warning or offered a transfer or pay cut instead
Basically, the last few years dealing with this industry have given me some trust issues
> I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week.
Government agencies/contractors for said agencies.
I've been messing with contrastive learning in the CLIP-like "minibatch" style for applications where correspondences between modalities within an example are a relevant source of information, and have been pleasantly surprised by how much this can actually in some cases speed up or otherwise aid with learning different objectives framed as decoding each modality's embedding for a different downstream objective. It seems that tying the representations together in this way can serve as a kind of semi-orthogonal correcting force on the embedding's ability to retain its original reconstructability while not constraining it too much to shift the representation for other goals, and many properties of the resulting models (like the obvious one of having interchangeable downstream decoding functionalities regardless of which modality is used to encode) suggest to me that it might be another tool to use when pursuing a better theory of objective balancing, as well as significant practical use in applications obviously. I'm surprised people aren't making more random contrastive embeddings for sets of "non-obvious" coupled representations
I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week. Does that even exist anywhere without like a PhD from rich people schools? Aside from the moral issues this myopic mindset of businesses just refusing to trust or support their people in any sense sure makes it hard to believe they're actually capable of "innovating" anything. Necessity may be the mother of invention, but focusing on speedrunning exploiting will always leave exploring neglected
> I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week.
Here's my tip: Get a regular mid-level web development job; spend 40% of the week to do the minimum of actual work you were allocated (just try not to dip below that), and spend the rest of your time doing whatever other work you want; tell your boss that while you're finding your allocated work difficult, this additional work that you're doing "out of hours" is really motivating you. The best case scenario is that they'll find that "extra" sufficiently useful to actually change your job title; worst case, you're let go, and can try again in a different company, having gained experience.
I mean I've worked for multiple companies wherein I've been explicitly hired as R&D staff and then let go because they don't want that anymore and would rather downsize than pay another dev to do the short-term work they are now focusing on, and also for companies where the workload wasn't too heavy and left me time to do other stuff but they ran out of runway without telling anyone. I've also worked for a company where I did three people's jobs that weren't mine for three months and got commended for saving the company (not my words) but then unceremoniously laid off because promises were made during that time (that I didn't ask for) and then spreadsheets were consulted and said promises started looking expensive
During none of these instances was I given any warning or offered a transfer or pay cut instead
Basically, the last few years dealing with this industry have given me some trust issues
> I really gotta find a job where they pay you but people won't freak out if you can't make a simple business case for every detail of what you're doing in a given week.
Government agencies/contractors for said agencies.