"<i>This form of deliberative reasoning is essentially ‘imagination’, it is a distinctly human ability</i>"<p>A completely unfounded supposition, as so often appears to be the case when some human monopoly is claimed. We didn't magically sprout whole new categories of ability during a measly few million years of evolution.<p>Anecdotally, I see crows getting out out the way of my car. Not confused and haphazardly as many birds do, but in calculated, deliberate, unhurried steps to somewhere <i>just</i> outside my trajectory - steps which clearly takes into account such elements as my speed and the state of other traffic on the road. Furthermore, when it's season for walnuts and the like, they'll calmly drop their haul on the asphalt, expecting my tyres to crush it for them. This - in my rural bit of Northern Europe - appears to be a recent import or invention; I never saw it done until two years ago.<p>And there's The Case of the Dog and the Peanut Butter Jars. My dog, my peanut butter jars, and they were empty, but not cleaned. Alone at home, she found them one day, and clearly had experimented on the first one, which had bitemarks aplenty on the lid. The rest she managed to unscrew without damage. Having licked the jars clean, apparently she got to thinking of the grumpy guy who woul eventually be coming home. I can think of no other explanation why I found the entire stash of licked-clean jars hidden - although not succesfully - under a rug.<p>Tell me again about imagination and its distinctly human nature.
<a href="https://en.wikipedia.org/wiki/Model_predictive_control" rel="nofollow">https://en.wikipedia.org/wiki/Model_predictive_control</a><p>Of course imagining possible outcomes before executing is useful! And it has many uses outside deep learning. No reason to reinvent new words, really. At least without referring to the established ones.<p>Maybe there <i>is</i> a serious novel idea, but I've missed it.<p>Basically, if you need to control a complex process (i.e. bring some future outcome in accordance to your plan), you can build a forward model of the system under control (which is simpler than a reverse model), and employ some optimization techniques (combinatorial, i.e. graph-based; numeric derivative-free, i.e. pattern-search; or differential) to find the optimal current action.
I'm not a planning guy but I work in a closely related community so I'm a least somewhat familar with the area.<p>Looking at the first paper (<a href="https://arxiv.org/pdf/1707.06170.pdf" rel="nofollow">https://arxiv.org/pdf/1707.06170.pdf</a>), it seems surprisingly shallow and light on details. So they have a learning system for continuous planning. So what? The AI Planning community has been doing this for ages with MDPs and POMDPs, solving problems where the planning domain has some discrete variables and some continuous variables. Here's a summary tutorial from Scott Sanner at ICAPS 2012:
<a href="http://icaps12.icaps-conference.org/planningschool/slides-Sanner.pdf" rel="nofollow">http://icaps12.icaps-conference.org/planningschool/slides-Sa...</a><p>Speaking of ICAPS: this conference is the primary venue for disseminating scientific results to researchers in the area. Yet the authors here cite exactly <i>one</i> ICAPS paper. WTF?<p>My bullshit detector is blaring.
The obvious caveat: this is quite far away from my field of expertise. Doubly so, because I'm not an expert in neural net ML and neither in cognitive science. So take this with spoonful of salt. But <i>anyhow</i>, I don't like the word "imagine" here. It seems suggest cognitive capabilities that their model probably does not have.<p>As far as I do understand the papers, their model builds (in unsupervised fashion which sounds very cool) an internal simulation of the agent's environment and runs it to evaluate different actions, so I can see why they'd call it imagination / planning, because that's the obvious inspiration for the model and so it sort of fits. But in common parlance, "imagination" [1] <i>also</i> means something that relatively conscious agents do, often with originality, and it does not seem that their models are yet that advanced.<p>I'm tempted to compare the choice of terminology to DeepDream, which is not exactly a replication of the mental states associated with human sleep, either.<p>[1] <a href="https://en.wikipedia.org/wiki/Imagination" rel="nofollow">https://en.wikipedia.org/wiki/Imagination</a>
Off topic: I posted this exact article four days ago: <a href="https://news.ycombinator.com/item?id=14813807" rel="nofollow">https://news.ycombinator.com/item?id=14813807</a><p>In the past, when I post exact duplicates, HN redirects me and automatically upvotes the original instead. I wonder why this doesn't always happen. (I'm not bothered, just curious.)<p>Double off topic: It's very interesting to see how much difference timing makes. My original had a single upvote, and this hit the front page.
I'm likely completely missing the point but how is this concept of imagination different from looking ahead in a search tree? Isn't exploring a search tree like in Chess or Go exploring future possibilities and their consequences before you decide on what to do next?
Why do we need to explicitly design architectures such as the "imagination encoder" the article describes? A proposed long term goal of deep learning is to have AI that surpasses human cognition (e.g. DeepMind's About page touts that they are "developing programs that can learn to solve any complex problem without needing to be taught how"), which was <i>not</i> explicitly designed in terms of architectural components such as an "imagination encoder".<p>Shouldn't imagination and planning be observed spontaneously as emergent properties of a sufficiently complex neural network? Conversely, if we have to explicitly account for these properties and come up with specific designs to emulate them, how do we know that we are on the right track to beyond human levels of cognition, and not just building "one-trick networks"?
> particularly in programs like AlphaGo, which use an ‘internal model’ to analyse how actions lead to future outcomes in order to to reason and plan.<p>I was under the impression that AlphaGo makes no plan but responds to the current board state with expert move probabilites that prunes MCTS random playouts.<p>There is no plan (AFAIK) or strategy in the AlphaGo papers so I find this statement that AlphaGo is an imaginative planner quite curious.<p>Perhaps someone can reconcile these statements or correct my knowledge of AlphaGo ?<p>Very interesting papers, it will be nice to see the imagination encoder methods applied to highly stochastic enviroments or indeed a robot in the real world.
Painful paper to read because of the inaccurate use of the word 'imagination'.<p>I'm sure the guys who wrote this are smart enough to know its not imagination (perhaps arguably a small subset of the attributes that contribute to what we know as imagination, but not imagination itself).<p>Which leads me to assume this hyperbole is there purely for the benefit of PR and stock price.
Evaluating different outcomes far ahead may be very computational intensive. One thing that AlphaGO shows is that a simple approach with Monte Carlo tree search can drastically cut down the search space. The "imagine" part could be just guided random walk ahead in planning, with something like Monte Carlo tree search.