Active Inference LiveStream 057.0 ~ Active Data Selection and Information Seeking

Background and Context video for:
Active Data Selection and Information Seeking
www.mdpi.com/1999-4893/17/3/118
Parr, Thomas, Karl Friston, and Peter Zeidman. 2024. "Active Data Selection and Information Seeking" Algorithms 17, no. 3: 118. doi.org/10.3390/a17030118
Abstract
Bayesian inference typically focuses upon two issues. The first is estimating the parameters of some model from data, and the second is quantifying the evidence for alternative hypotheses-formulated as alternative models. This paper focuses upon a third issue. Our interest is in the selection of data-either through sampling subsets of data from a large dataset or through optimising experimental design-based upon the models we have of how those data are generated. Optimising data-selection ensures we can achieve good inference with fewer data, saving on computational and experimental costs. This paper aims to unpack the principles of active sampling of data by drawing from neurobiological research on animal exploration and from the theory of optimal experimental design. We offer an overview of the salient points from these fields and illustrate their application in simple toy examples, ranging from function approximation with basis sets to inference about processes that evolve over time. Finally, we consider how this approach to data selection could be applied to the design of (Bayes-adaptive) clinical trials.
Keywords: experimental design; active sampling; information gain; Bayesian inference
Active Inference Institute information:
Website: activeinference.org/
Twitter: / inferenceactive
Discord: / discord
KZbin: / activeinference
Active Inference Livestreams: coda.io/@active-inference-ins...

[@syntheticdatagenerator]

Daniel - awesome discussion. Great insights on data as information gain opportunities.

[@margrietoregan828]

Define your terms ! ‘Data’ ? ‘Information’ ? Although you folk at Active Inference Institute use the term ‘data’, this term is just techie speak for ‘information’. So, what is the definition of ‘information’ ? INFORMATION. What is ‘information’ - as a phenomenon in its own right & not just what any of it ‘says’ or means or does. ‘Thought’, ‘mind’, ‘intelligence’, ‘cognition’ & ‘consciousness’ are all information-related phenomena and it is not difficult to show that one of the principal reasons we have not so far come to any good, proper & fully verifiable understanding of these otherwise greatly sought-after yet still highly mysterious phenomena is due in great part to the simple fact that we do not presently also have a good & proper, fully verifiable understanding of ‘information’ itself. Although I have personally had the (dubious) fortune of having been able to figure out ‘information’s’ correct (& fully verifiable) ontological identity - plus a full science of the phenomenon to boot - and although I’m not going to divulge its formalistic definition here in this KZbin comment (without which formalistic definition it is not possible to establish a full & accurate science of the phenomenon, but with it it is) nevertheless I can assure you that with it in hand - that is, with ‘information’s’ correct ontological identity within one’s investigative arsenal (along with a full science thereof also), the exercise of determining the ontological identities of all of the other directly information-related phenomena (along with an accompanying science of each also) such as ‘thought’, ‘mind’, ‘cognition’, ‘intelligence’, ‘learning’, ‘understanding’, ‘sentience’, ‘consciousness’ & ‘self-consciousness’ (to far less than exhaust the list) becomes one of no great difficulty. A full science of any one particular phenomenon will include understandings & quantifications of such things as its nature (its ontological identity), its standing (in the existential hierarchy), its role, function, varieties, distribution, incidences, properties/extensions/capacities/capabilities, usages, handling- &/or operational amenabilities, & its typical life history (its creation through to its dissolution/erasure). Indeed, once both ‘information’s’ correct ontological identity is known along with a full science thereof, much clarifying light is thrown upon essentially all of the other elements of reality. Nevertheless & however surprising - given that we live smack bang right in the middle of ‘the information age’ - the simple, easily verified fact of the matter is, nowhere within the annals of the entire compendium of human discourse, is ‘information’ defined with any rigour - let alone with sufficient rigour as would allow it to be included in any properly scientific - or philosophical - research or work. …. Yet, just as individual units of the pictographic & acoustic information going through any (analogue) movie-making-&-showing process (from set to big screen) can be readily identified, observed, followed & understood, so too can the individual units of all of the different kinds of information going through our brains be tracked-&-traced also. This task becomes as unproblematic as it is once ‘information’s’ correct identity is known & understood - including the facts that ‘information’ is not computer digits, & computing is not thinking. Furthermore, tracking-&-tracing the movement of all of the streams of information/data which eventually make up the content of our own consciousness, enables the exact location of the latter to be determined, & which Mark Solms has already determined as being within the periaqueductal gray region of the brain stem …. Indeed, examining the particular component of our brain which exists at this particular ’Mark Solms spot’ further allows a determination of exactly what consciousness itself is - as a phenomenon in its own right - to be made……. Voila !

[@ActiveInference]

Thank you for sharing that, many interesting topics and claims. Looking forward to learning more however is best.