- Born
- 2000
- Field
- researcher
- In the library
- 3 works, named in 1 passages elsewhere
Energy and information
In a 2013 synthesis with Stemmler and Friston, Sengupta and colleagues linked the energy used by neurons to information. Signalling changes the number of molecules in fixed-volume compartments, and the authors note that the Nernst potential arises from this. For ions such as sodium and potassium the changes are small, so linearising the work term gives the energetic cost of signals1. The paper then casts organisms as systems that avoid phase transitions by bounding the entropy of their sensory states2. Exploration is not a paradox here, because prior beliefs can include it: agents expect to explore3.
Cells and pattern regulation
The 2015 paper, with Friston, Levin and Pezzulo, applies the same principle to morphogenesis. Its premise is that each cell has an implicit model of its place in the final form, and that self-assembly means sampling the signals that model predicts4. The authors frame this as minimisation of variational free energy. They describe internal and active states as following a generalized gradient ascent on an ergodic density, which they tie to thermodynamic free energy in non-equilibrium steady states5.
Gauge theory and later use
In the 2016 paper, Sengupta and co-authors ask whether variational free energy, if it is ubiquitous in living systems, could serve as the Lagrangian of a gauge theory6. They say such a formulation might also bear on new inference schemes for data analysis, but they focus on neurobiology7. Levin, in 2019, cites this paper among work on geometric information theory. He suggests the analogy between geometric and cognitive spaces needs further study8.
By linearising δW in the concentration changes, we can easily compute the energetic cost of neuronal signals [7].
Sengupta B, Stemmler MB, Friston KJ, 2013 · Information and efficiency in the nervous system--a synthesis · open at passage 4The free-energy principle suggests that organisms avoid phase transitions by minimising (a variational free energy bound on) the Shannon entropy of their sensory states.
Sengupta B, Stemmler MB, Friston KJ, 2013 · Information and efficiency in the nervous system--a synthesis · open at passage 17In other words, agents expect to explore and would be surprised if they did not.
Sengupta B, Stemmler MB, Friston KJ, 2013 · Information and efficiency in the nervous system--a synthesis · open at passage 17we pursue the notion that morphogenetic self-organization requires each cell to have an implicit model of its place in the final morphology [21]
Friston K, Levin M, Sengupta B, Pezzulo G, 2015 · Knowing one's place: a free-energy approach to pattern regulation · open at passage 4It says that their flow performs a generalized gradient ascent on the marginal ergodic density over internal states and their Markov blanket denoted by m.
Friston K, Levin M, Sengupta B, Pezzulo G, 2015 · Knowing one's place: a free-energy approach to pattern regulation · open at passage 14If the minimisation of variational free energy is a ubiquitous aspect of biological systems, could it be the Lagrangian of a gauge theory?
Sengupta B, Tozzi A, Cooray GK, Douglas PK, Douglas PK…, 2016 · Towards a Neuronal Gauge Theory · open at passage 12Although formulating the free energy principle as a gauge theory may have important implications for new inference schemes that can be used to analyse data
Sengupta B, Tozzi A, Cooray GK, Douglas PK, Douglas PK…, 2016 · Towards a Neuronal Gauge Theory · open at passage 4the analogies between geometric spaces and cognitive ones needs to be explored further, as is beginning to be done via geometric information theory (Balduzzi and Tononi, 2009; Oizumi et al., 2016; Sengupta et al., 2016; Calcagni, 2018)
Levin M, 2019 · The Computational Boundary of a "Self": Developmental Bioelectricity Drives… · open at passage 77
2013
2013 · read
2015
2015 · read
2016
2016 · read