I'm looking for studying resources on memory models for storage and recall of sequences, particularly associative and auto-associative recall of such sequences. Ideally I would like a textbook on this with exercises at the ends of the chapters and actual sourcecode. I actually want to learn the topic towards the aim of implementation.
(These might be Boltzmann Machines, Hopfield Networks, or SOMs, but it is unclear whether these were ever used for sequences.)
Russell and Norvig's AIMA does not cover this topic in any level of detail that I need, being a historical survey text. I am primarily focused in models that are not trained by backprop, but instead rely upon local changes to synaptic weights in a unsupervised manner. Although I don't mind reading some chapters on supervised learning methods too. Thanks.
To give a better idea of the material I am wanting to dig into , I give the two following abstracts, both of which are from the 1980s. (Yes, the 1980s). I'm hoping there is some updated textbooks and materials on this topic.
Associative Memory in a Simple Model of Oscillating Cortex (1990)
> The network has explicit excitatory neurons with local inhibitory interneuron feedback that forms a set of nonlinear oscillators coupled only by long range excitatofy connections. Using a local Hebb-like learning rule for primary and higher order synapses at the ends of the long range connections, the system learns to store the kinds of oscillation amplitude patterns observed in olfactory and visual cortex. This rule is derived from a more general "projection algorithm" for recurrent analog networks, that analytically guarantees content addressable memory storage of continuous periodic sequences - capacity: N /2 Fourier components for an N node network - no "spurious" attractors.
.
.
Adaptive bidirectional associative memories (1987)
> Two fields of neurons, FA and FB, are connected by an n X p synaptic marix M. Passing information through M gives one direction, passing information through its transpose M^T gives the other. Every matrix is bidirectionally stable for bivalent and for continuous neurons. Paired data (Ai,Bi) are encoded in M by summing bipolar correlation matrices. The bidirectional associative memory (BAM) behaves as a two-layer hierarchy of symmetrically connected neurons. When the neurons in FA and FB are activated, the network quickly evolves to a stable state of twopattern reverberation, or pseud
... keep reading on reddit ➡HYPOTHESIS AND THEORY article
Front. Mol. Neurosci., 25 February 2021 | https://doi.org/10.3389/fnmol.2021.592951
The Mechanical Basis of Memory – the MeshCODE Theory
- School of Biosciences, University of Kent, Canterbury, United Kingdom
One of the major unsolved mysteries of biological science concerns the question of where and in what form information is stored in the brain. I propose that memory is stored in the brain in a mechanically encoded binary format written into the conformations of proteins found in the cell-extracellular matrix (ECM) adhesions that organise each and every synapse. The MeshCODE framework outlined here represents a unifying theory of data storage in animals, providing read-write storage of both dynamic and persistent information in a binary format. Mechanosensitive proteins that contain force-dependent switches can store information persistently, which can be written or updated using small changes in mechanical force. These mechanosensitive proteins, such as talin, scaffold each synapse, creating a meshwork of switches that together form a code, the so-called MeshCODE. Large signalling complexes assemble on these scaffolds as a function of the switch patterns and these complexes would both stabilise the patterns and coordinate synaptic regulators to dynamically tune synaptic activity. Synaptic transmission and action potential spike trains would operate the cytoskeletal machinery to write and update the synaptic MeshCODEs, thereby propagating this coding throughout the organism. Based on established biophysical principles, such a mechanical basis for memory would provide a physical location for data storage in the brain, with the binary patterns, encoded in the information-storing mechanosensitive molecules in the synaptic scaffolds, and the complexes that form on them, representing the physical location of engrams. Furthermore, the conversion and storage of sensory and temporal inputs into a binary format would constitute an addressable read-write memory system, supporting the view of the mind as an organic supercomputer.
I have to change a bunch of lb vservers on Citrix ADC to non addressable (IP:0.0.0.0 Port:0) and put them behind content switches. There are many so I'd like to do it from CLI, or ADM config job. However it doesn't seem to work and I swear I figured this out a year or two ago. Does anyone know the command to change the IP and Port of a regular LB vserver to non addressable from CLI? The command to add one is "add lb vserver tst_nonaddr SSL 0.0.0.0 0 -persistenceType NONE -cltTimeout 180" but no version of the "set" command to change to this seems to work.
Like the title suggests. I'm thinking of setting up a network share to load PS2 games via OPL. What is the largest pool of storage OPL can handle? Is it just whatever the max of the filesystems it can handle?
It is a joke that NVAX would trade down on the SCOTUS news - that means nothing. The W.H.O. had a goal of vaccinating 40% percent of the population in every country by YE 2021 and they did not hit the goal. The new goal is 70% of population in every country by mid 2022. The lowest rates of vaccination are in the developing world where NVAX will have much greater uptake because it can use traditional cold chain storage. In the developing world we are looking at a annual shot or booster. The groups still to be vaccinated are kids (data suggests kids have less reactions to NVAX) and some anti-vax types may convert with a non-mRna - traditional protein based vaccine. The addressable market is every person on earth. COVID is here to stay and it will be an arms race between vaccine manufacturers and the Virus Variants. NVAX will start leading that battle and we should get behind them. (Disclaimer - Despite getting crushed yesterday. I am all in on NVAX $240K in stock and Feb, Mar and April call options today worth $210k total today). The only real risk is if they run into manufacturing or quality issues. They have had some time to address these so I think it is safe to believe the will really ramp up and deliver. The stock is at a 52 week low and the news flow will be very strong for the next 90 days. Consensus price tragets call for a double..... don't see opportunities like this often.
https://preview.redd.it/u4mp22e3uvb81.png?width=2524&format=png&auto=webp&s=0198684fd1bca7b29f39a46bc3c1d40d8e9b2789
TLDR: I bought a xunlai storage page and a extra char slot, but after a bizarre bug redeeming the codes, only the char slot was added to account. Yes, I have triple-checked that I did not purchase the material storage expansion by accident. Support process is in progress, but currently haven't received a working code, and after two replies I am not sure I will
Apologies for grammar, english is not my primary language. I did pre-check this post with mods for not breaking the "Contacting Support"-rule. Skip down to "Main point of this post" if you don't want to read the background or my experience with support
Background: So after purchasing xunlai page and char slot 31st of december, I applied both codes I got pretty much instantly in-game throught the "manage access keys". I remember that I added the xunlai code first, which correctly increased the amount of pages I had unlocked under the "extra features" header on that page. Then I added char slot code, and saw my char slot amount increase by two and the xunlai page decrease by one. I thought this was maybe some visual bug or maybe I was just seeing things. After logging out and relogging, I had the correct amount of char slots, but even as of now, the xunlai page has not appeared in-game. Both codes are also registered as used.
Support process in a nutshell: I sent in a short ticket describing that I had a purchase where one of the codes simply didn't work, and described the issue. I received an answer that they could clearly see I had "the item "Xunlai Storage Pane" & "Extra Char Slots" is already registered to your account.". I clarified the issue again, and asked them to rectify the problem by unlocking the content I paid for but did not receive. I got an answer back that due to the age of the game, support is limited to account and payment issues only. I haven't replied back yet, but since this is pretty clearly and payment issue in my mind, I'll try to ask them to send me a code that works. However, I am mostly considering the amount of 6,99 eur already lost, and going through the support process just because of the principle
Main point of this post: I am making this post, because it seems like it is possible, even if very unlikely, to receive a access key that doesn't work. From this experience it looks like support might be unable to fix this situation (where the code "works" from being able to redeem it, but doesn't actually unlock anything in-game). Of course it might just be that I com
... keep reading on reddit ➡I'm so upset. I have just about 70 hours in this game and only started playing a few weeks ago. I have a bunch of very organized storage boxes and every in-game night I empty my pockets into my boxes. It was like 1:50 am so I was just finishing up and at 1:58 I accidentally pressed clear away one of boxes and everything landed on the ground and the game reset and autosaved.
It was my gem/jewelry box. I lost all my raw gems and processed gems (about 30 total for each kind), all my olive crystals, all my jewelry from my pond draining (necklaces, rings, watches etc) everything I've been grinding for for 6 seasons in the game.
Why is the storage system so bad? :'(
Any progress since the hopfield network? Content addressable memory seems very important concept , couldn't find anything similar in deep learning. Anything?
I absolutely love the base game and I definitely want to get the expansions. However, carrying around everything might be a bit of a hassle and wanted to check and see if anyone has been able to find a case or bag that could fit all the miniatures and deck boxes containing sleeved cards. I want to be able to bring the game with me to friends' houses but bringing individual boxes and components almost guarantees me losing something along the way.
My first thought was to search for Warhammer cases but they all seem to only accommodate smaller figures. Any help and suggestions are welcome! :)
txtai 4.0 has been released with a number of new features.
https://preview.redd.it/qm9tgjkum2b81.png?width=1061&format=png&auto=webp&s=22c90382d1dee780530e69635242078a26b8305c
- Content Storage - Content can now be stored alongside embeddings vectors. No longer required to have external data storage.
- Query with SQL - txtai supports both natural language queries and SQL queries
embeddings.search("feel good story")
SELECT id, text, score FROM txtai WHERE similar('feel good story') AND score >= 0.15 - Object Storage - Store binary objects alongside embeddings vectors
- Reindex - Indexes can be rebuilt using stored content, no need to resend data to txtai
- Index Compression - Indexes can be compressed using GZ/XZ/BZ2/ZIP
- External Vectors - Use external vector models from an API or an external library. Centralize building vectors on GPU servers leaving index servers to be powered by more modest hardware.
More information can be found in following links.
